Wage Incentives as a Managerial Tool 9780231899352

Citation preview

WAGE INCENTIVES as a Managerial Tool

WAGE INCENTIVES as a Managerial Tool by WILLIAM B. WOLF

1957 Columbia University Press, New York

Copyright © 1957 Columbia University Press, New York Published in Creat Britain, Canada, India, and Pakistan by the O x f o r d University Press London, T o r o n t o , Bombay, and Karachi Library of Congress Catalog Card Number: 57-5939 Manufactured in the United States of America

To Nancy, Pete, and Steve

Preface T h i s is a study of the use of wage incentives in manufacturing plants. It was originally started in 1944. At that time I was in charge of the standards department of a relatively large company. In comparing my experience in administering wage incentives with the reports appearing in the current literature I noticed significant discrepancies between expectations and results. Furthermore, talks with fellow practitioners failed to provide a logical framework for viewing wage incentives. T h u s this study was started in order to discover the sense and/or non-sense of using wage incentives to remunerate rank-and-file factory workers. T h e particular contribution which it purports to make is to give both the student and the practitioner a fuller understanding of the use of this technique. In the preparation of this book I am deeply indebted to many businessmen and union leaders. T h e y gave patiently of their time and knowledge to supply much of the data in this book. However, for obvious reasons they must remain anonymous. A number of my colleagues have given helpful suggestions, especially Professors William Vatter, Raleigh W . Stone, Robert K. Burns, and Ezra Solomon, of the University of Chicago, and Professors Cornelius Gillam and Robert A. Sutermeister, of the University of Washington. In addition, the support of Dean Austin Grimshaw of the College of Business Administration of the University of Washington has done much to make this book a reality. WILLIAM The College of Business Administration The University of Washington Seattle, Washington October, 1956

B.

WOLF

Contents I.

THE

N A T U R E OF W A G E

Introduction

INCENTIVES

3

3

Definition of W a g e Incentives

5

W a g e Incentives as a Managerial T o o l T h e Development of W a g e Incentives Wage Incentive Formulae II.

THE

IMPACT

COSTS:

AN

OF

WAGE

EMPIRICAL

Introduction

6 8

10

INCENTIVES STUDY

ON

MANUFACTURING

17

17

Scope and method of the empirical investigation Character of the

findings

Presentation of the

findings

17

19 20

T h e Effect of W a g e Incentives on the General Level of Plant Wages

21

Operating Problems Associated with the Use of W a g e Incentives 24 Setting standards of output 25 Maintaining an acceptable wage structure 30 Management Activities Associated with the Administration of Wage Incentives 48 Explaining the wage incentive system to employees 48 Setting and maintaining sound standards 57 Installing the standard 58 Avoiding or correcting inequities in the wage structure 62 Preventing fraudulent practices 66 A n Evaluation of the Impact of Wage Incentives on Costs 68

Contents

X III.

THE A

T H E O R Y AND L O G I C REEXAMINATION

OF W A G E

INCENTIVES:

72

T h e Theory of Wage Incentives

72

T h e Logic of the Use of Wage Incentives in the Manufacturing Plant 73 T h e nature of the manufacturing plant 73 T h e contribution of wage incentives to the manufacturing plant's objective 75 Application of the wage incentive theory to manufacturing plants 77 Implications of the Logic of Wage Incentives under Operating Conditions 79 T h e determination of uniform and meaningful measures of production 79 T h e determination of standards of performance 80 Wage incentives as a device for motivating workers to increase their rate of output 85 Effect of increased output on marginal cost 91 Wage incentives as a device for stimulating manage93 ment to be more efficient IV.

UNDERSTANDING THE U S E O F W A G E INCENTIVES

114

T h e Importance of Wage Incentives in Plant Management 114 Precepts and Caveats Conclusion BIBLIOGRAPHY INDEX

139

127 131

118

WAGE INCENTIVES as a Managerial Tool

I

The Nature of WAGE

INCENTIVES

INTRODUCTION

In spite of the attention that the Scientific Management Movement and the Second World War emergency focused on the use of wage incentives, very little has been done to analyze their implications for management. 1 In the management literature, most of the attention has been centered on descriptions of specific wage incentive plans, discussion of the ostensible benefits of specific plans, and mechanical comparisons of the cost curves associated with various wage incentive formulae. 2 1 One of the first tasks in the preparation of this study was to review the existing literature on wage incentives. T h e review was made to determine what had already been done and to provide a check for the findings of this investigation. Though the review revealed a vast amount of writing, most of it is from a nonmanagement point of view. For example, it appears that Karl Marx was one of the first writers to study the impact of wage incentives systematically. He analyzed early English experience as revealed by the official reports of the Inspectors of Factories and the Children's Employment Commission. See Capital (Modern Library ed.; New York, Random House, Inc., igo6), pp. 6 0 5 - 1 1 . For other early comments on the use of wage incentives, see Adam Smith, Wealth of Nations (Modern Library ed.; New York, R a n dom House, Inc., 1937), Book I, Chap. VIII, p. 61; Sidney and Beatrice Webb, Industrial Democracy (London, Longmans, Green and Co., 1902), Chap. V. T h e most thorough studies of wage incentives have dealt with union attitudes and activities. For example, the Webbs, Industrial Democracy; David A. McCabe, " T h e Standard Rate in American Trade Unions," in Johns Hopkins University Studies in Historical and Political Science (Baltimore. T h e Johns Hopkins Press, 1912), Series X X I , No. 2; G. D. H. Cole, The Payment of Wages (London, George Allen and Unwin, Ltd., 1918); Sumner H. Slichter, Union Policies and Industrial Management (Washington, Brookings Institution, 1941), Chaps. X and X I ; Van Dusen Kennedy, Union Policy and Incentive Wage Methods (New York, Columbia University Press, 1945). 2 Most of the management literature appears in periodicals. Very few books on wage incentives per se have been written. Those that do devote

4

Nature of Wage Incentives

An analysis of the classic book on the subject (Charles W. Lytle's Wage Incentive Methods) reveals scarcely a paragraph indicating the concepts underlying the use of wage incentives or the multiple complications arising from their use. This concentration on specific techniques and practices has lost sight of the general problem and has failed to induce formulation of the major concepts necessary for a proper perspective on the subject. It needs to be supplemented by an empirical study of the actual impact of wage incentives and an analysis of how these findings compare with the logic underlying their use. Such an analysis is necessary to provide both the student and the practitioner with a concise set of principles for dealing with wage incentives in their true perspective. This study is designed to provide such an analysis. Chapter II describes the impact of wage incentives as revealed by field investigations, Chapter III analyzes the logic of wage incentives, and Chapter IV sets forth precepts and caveats to guide those concerned with the use of wage incentives. Before the impact of wage incentives can be described, attention must be given to the definition of the term and to the nature of the use of this device. Accordingly, the present chapter discusses the definition of wage incentives, their use as a management technique, and the history of considerable attention to wage incentives are essentially descriptive and nontheoretical. For example: David F. Schloss, Methods of Industrial Remuneration (Oxford, England, Williams and Norgate, 3d ed. 1898); Daniel Bloomfield, Financial Incentives for Employees and Executives (New York, T h e H. W. Wilson Co., 1923), Vol. I.; L. P. Alford, ed.. Management's Handbook (New York, Ronald Press Co., 1924), Sec. 16.; Charles W. Lytle, Wage Incentive Methods (rev. ed.; New York, Ronald Press Co., 1938); Hugo Diemer, ed., Wage-Payment Plans That Reduce Production Costs (New York, McGraw-Hill Publishing Co., Inc., 1929); Carl C. Harrington, ed.. Job Evaluation and Wage Incentives (New York, Conover-Mast Publications, Inc., 1949); Norman C. Hunt, Methods of Wage Payment in British Industry (London, Sir Isaac Pitman and Sons, Ltd., 1951).

N a t u r e of W a g e Incentives

5

their development. It concludes with a brief description of some of the more common wage incentive plans. DEFINITION

OF W A G E

INCENTIVES

As used in this study, the term "wage incentives" refers to systems of remunerating rank-and-file workers under which the earnings of a worker, or a group of workers, are directly, promptly, and automatically related to his output by a predetermined formula relating his actual performance to a specific standard of performance. T h e essential characteristics of wage incentives are: (i) a standard of performance for each j o b or task is specifically established; and (2) the worker's earnings are directly, promptly, and automatically varied according to an established formula for relating actual performance to the standard. Wage incentives are to be differentiated from time wages.3 Under the latter method of wage payment, the worker is paid on the basis of the time he works rather than according to his output. T h i s difference is essentially one of degree. 4 In a time wage system, standards of output are usually implied, and frequently the worker's hourly wage rate reflects the level of his efficiency. However, the standard of output is implicit rather than explicit, and the relationship between earnings and output is not definite, automatic, or immediate. Frequently time wages and wage incentives are combined within a single wage payment system. Workers may s In the literature there is a great deal of confusion in terminology. W h a t i j referred to above as "time wages" is also sometimes called "day work," "time work," or "straight-time." * T h e line of demarcation between wage incentives and time wages is faint. For example: Under a measured day rate system, explicit standards of performance are determined. However, if no definite and precise relationship between remuneration and o u t p u t is established, the system is not technically a wage incentive. If there is a definite and precise method for promptly relating the worker's output to his earnings, the system is classified as a wage incentive.

6

Nature of Wage Incentives

be given a guaranteed minimum rate of earnings per hour and then be paid a bonus 5 or premium for all production in excess of the established standard of output. Such combinations are classified as wage incentives because the level of incentive earnings is usually considerably above the hourly guaranteed wage rate, and the characteristics of wage incentives are present (that is, explicit standards of performance are set, and, within limits, the relationship between output and earnings is definite, immediate, and automatic). WAGE

INCENTIVES AS A

MANAGERIAL

TOOL

Wage incentives are a managerial tool. Their primary purpose is to aid in obtaining minimum unit costs, thereby contributing to enterprise profits. Lytle says that "the primary and universal reason for the installation of wage payment plans (i.e., wage incentives) is today, as ever, to secure the lowering of unit costs on the one hand, and to improve the earnings of the employees on the other." 6 T h e manner in which wage incentives reduce costs has been described in terms of numerous intermediate objectives. For example, a representative of the Westinghouse Electric Elevator Company stated that the objectives and anticipated results of a scientifically worked out wage incentive plan are: 1. Production problems or maladjustments are brought to immediate attention due to a drop in the incentive bonus factor. 2. Better segregation of efficient workers thus insuring a full utilization of available manpower. 3. Better planning of details of the job and also the methods of the job by supervisors or industrial engineers. 8 As used in this study, the term " b o n u s " refers to earnings in excess of the worker's guaranteed hourly wage rate. • L y t l e , Wage Incentive Methods, p. 53.

Nature of W a g e Incentives

7

4. Increase in production due to the accumulation of the above benefits plus the major factor of increased effort on the part of the operator. 5. Better morale of the worker whenever the plan is administered efficiently and fairly. a. Because of his effort, he can increase his earnings. b. By his production, he can gauge his own merit and work. c. By seeing immediate return for efforts expended, he applies himself to his task willingly and cheerfully. 7 R i e g e l 8 lists several slightly different benefits which result f r o m the use of wage incentives: 1. Greater output per hour on each job on incentive brought about lower unit costs for direct labor and overhead. 2. Accurate predetermination of costs—essential in mass production industries which contract for production in advance—was made possible. 3. Unit costs were stabilized while wage rates and prices of material remained constant. 4. T h e reduction of watching and bossing to obtain employee application on the job had some favorable influence on industrial relations. 5. T h e production standards, which are apart from but essential to an incentive plan, were helpful in planning, scheduling, and coordinating production, and in forecasting equipment and manpower needs. Lionel Michael enumerates the f o l l o w i n g advantages derived from " a sound wage incentive plan, properly conceived, installed and administered": 9 1. More effective operation methods and costs.

through

standardization

of

''Experience of 12) Companies with Wage Incentive Plans, Section I (Chicago, Oartnell Corporation, 1944), p. 12. s John W. Riegel, "Essentials in Incentive Compensation," in Developments in Wage Payment Techniques, American Management Association Personnel Series, No. 77 (1944), p. 12. 9 Lionel B. Michael, Wage and Salary Fundamentals and Procedures (New York, McGraw-Hill, 1950), pp. 235-36.

8

Nature of Wage Incentives

2. Increased production through improved methods and greater worker productivity. 3. Reduction of unit costs through greater utilization oí equipment and worker effort. 4. Increased market potentialities through providing a product at an attractive price. In summary, wage incentives are essentially a managerial tool for reducing cost. In achieving this goal, two closely related areas are emphasized: (1) the role of wage incentives in motivating workers to increase their rate of output, and (2) the role of wage incentives in stimulating management to be more efficient in performing its functions. T h e impact of wage incentives is to be appraised in terms of their efficacy in achieving these ends. However, before considering the actual impact of wage incentives, it is desirable to reflect upon the history of their development and related matters, thus providing a more thorough and complete background for the analysis of their use. THE

DEVELOPMENT OF W A G E

INCENTIVES

Wage incentives have a long history. T h e y have existed since the inception of the employer-employee relationship.10 Records show that the principal type of wage incentive—piece rate—was used in the ancient Egyptian, Roman, and Grecian economies. 11 One of the early uses of wage incentives was under the home-work system of manufacturing. Under this form of organization wage incentive problems differed from those in modern manufacturing. T h e incentive aspect of the wage payment system was probably of secondary importance, since the employer was interested primarily in receiving a specified quantity of work for a given price. As long as an adequate supply of labor was available, he was T h e Webbs, Industrial Democracy, p. 285. 11 Harlow S. Person, "Methods of Remuneration," in Encyclopedia of the Social Sciences, VIII (1935), 672-82. 10

N a t u r e of W a g e

Incentives

9

under little pressure to be seriously concerned about the worker's rate of output. T h e modern use of wage incentives is associated with the development of the factory system. With the centering of production in the factory the incentive aspect gained added importance, since the employer was now faced with the task of spreading his overhead by increasing output. T h e rapid industrialization during the last half of the nineteenth century directed increased attention to the limitations of wage incentives. 12 It stimulated the search for some form of wage payment which would have incentive value yet still be free from rate cutting and restriction of output. T h i s search led to innovations in wage incentive systems as well as to the consideration of various alternative means of motivating workers. 13 T h e development of modern wage incentive plans followed two general approaches. T h e first was the development of systems wherein employer and employees would share the savings in direct labor costs resulting from increased production. T h e logic underlying these plans was that the sharing feature would eliminate management's need and desire to cut rates. T h e Halsey and Rowan plans are examples of this approach. 14 T h e y provided for standards of output based on records of past performance, and 1 2 T h e point of emphasis was that workers' fear of rate cutting prompted them to restrict output. is Consideration of the alternatives to wage incentives is outside the scope of the present discussion. However, it is appropriate to point out that various forms of profit sharing and the use of nonfinancial incentives were suggested as solutions to the problems of rate cutting and restriction of output. See: Henry R. T o w n e , " G a i n Sharing," in Transactions of the American Society of Mechanical Engineers (New York, 1889), Vol. X; Nicholas P. Gilman, Profit Sharing between Employer and Employee (New York, Houghton Mifflin and Co., 1889); R . B . Wolf, "Nonfinancial Incentives," in Publications of the American Society of Mechanical Engineers (1918), N o . 1673, quoted in John R. Commons, ed.. Trade Unionism and Labor Problems (New York, Ginn and Co., 1921), pp. 218-32. 1 4 F. A . Halsey, " T h e Premium Plan of Paying for Labor," in Transactions of the American Society of Mechanical Engineers (New York, 1891), Vol. X I I ; W . Rowan T h o m s o n , The Premium Bonus System (Glasgow, Scotland, McCorquodale and Co., Ltd., 1917).

io

Nature of Wage Incentives

they paid workers only a part of the savings in direct labor cost arising from production in excess of the established standard. Since these plans provided a formula for relating wages to output in a manner that w o u l d eliminate the pressure for rate cutting, it was argued that they removed the main cause for restriction of output. T h e second approach to the p r o b l e m was a direct attack on the method of setting standards. T h e underlying reasoning was that setting correct standards of performance would automatically eliminate rate cutting and restriction of output. T h i s logic was first popularized by F. W . T a y lor. 10 His central theme is represented by the following quotation: Since [under the Taylor system] the rate-fixing is done from accurate knowledge instead of more or less by guess-work, the motive for holding back on work, for 'soldiering' and endeavoring to deceive the employer as to the time required to do the work, is entirely removed, and with it the greatest cause for hard feelings and war between the management and the men. 16 T h e procedure advocated by T a y l o r was "to resolve each j o b into its elements, to make a careful study of the quickest time in which each of the elementary operations can be done, and then to properly classify, tabulate, and index this information, and use it w h e n required for rate fixing." "

WAGE

INCENTIVE

FORMULAE

18

From 1891 to the 1920s, numerous formulae for relating earnings to output were developed. Some had u n i q u e fea1 5 F. W . T a y l o r , " A Piece-Rate System." in Transactions of the American Society of Mechanical Engineers (New York, 1894 and 1895), X V I , 856-903. '«ibid., pp. 857-58. " Ibid., p! 869. Since this study is concerned with the more general aspects of wage incentives, only cursory attention is given to the specific formulae. For de-

Nature of Wage Incentives

xi

tures; others were simply identical plans under different names. T h i s section describes only a few of the better known plans. These are the piece rate plan, the standard hours plan, the Halsey Premium Plan, the Rowan Plan, the Bedaux Point Plan, the Taylor Differential Piece Rate Plan, the Merrick Multiple Piece Rate, the Gantt Task and Bonus System, and the Parkhurst Differential Bonus System. Piece rate. T h e piece rate system is the oldest form of wage incentive. It provides for payment to workers at a constant price per unit of output. T h e worker's earnings are determined by multiplying his output by the price per unit. Under most modern piece rate plans, a minimum hourly rate of pay is guaranteed to the worker. Hence, for low levels of output, the plan becomes a time wage plan. Standard hours plan. T h e only significant difference between the standard hours plan and the piece rate plan is that for the former the standards of performance are quoted in units of time, rather than in money. Under a standard hours plan the worker is paid at his established hourly wage for all standard hours of work he produces, regardless of the time he actually takes to complete the work. For example, if a worker completes a job in four hours, and if six standard hours are allowed for that job, his earnings for the job will be six times his hourly rate of pay. Halsey Premium Plan. T h e Halsey Premium Plan is closely related to the standard hours plan. In fact, one vertailed discussion of the technical aspects of different wage incentive formulae, see: Lytle, Wage Incentive Methods; Diemer, Wage-Payment Plans That Reduce Production Costs, Parts I and II; National Industrial Conference Board, Systems of Wage Payment (New York, 1930); International Labour Office, Payment by Results (Geneva, 1951); L. P. Alford and John R. Bangs, eds., Production Handbook (New York, R o n a l d Press Co., 1944), Sec. 18.

12

Nature of Wage Incentives

sion of the Halsey plan (the 100 percent premium plan) is identical with the standard hours plan. 19 The more common version of the Halsey plan provides that employer and employees share the savings in direct labor cost arising from above-standard output. In the United States, the exact percentage of the savings going to labor varies. T h e most usual sharing is on a 50-50 basis. An example will serve to illustrate the mechanics of the Halsey 50-50 sharing plan. If six standard hours are allowed for a job and the worker actually performs the job in four hours, his earnings are calculated as follows: E = T R + (A -

T)PR

Where E equals earnings, T equals time taken, A equals time allowed, R equals the worker's hourly rate of pay, and P equals the percentage of time saved paid to the worker (in this case, 50 percent). In this example earnings are equal to four times the worker's hourly rate of pay plus his hourly rate times one-half of the time saved. E = 4R + (6 - 4) , 5 R E = 4R + R E = 5R As pointed out above, this plan was specifically designed to reduce the need for rate cutting. Halsey reasoned that allowing management to share directly in the savings of direct labor cost mitigated management's need to cut rates and so eliminated the main cause for restriction of output. Rowan Plan. The Rowan Plan is similar to the Halsey sharing plan in that it, too, provides a system whereby management and labor may share the savings of direct 19 It should be emphasized that this discussion is limited to the f o r m u l a e f o r relating earnings to output. As originally described by Halsey, his plan was designed for use with loose standards of performance based on past performance. As used in present-day industry this is not necessarily the case. T h e Halsey p l a n often is used in connection with standards of output determined by systematic time and motion study.

N a t u r e of Wage Incentives

13

labor costs resulting from above-standard performance. T h e significant difference between the Halsey plan and the Rowan Plan is that under the former the sharing of savings between workers and employer is constant, whereas under the latter the workers receive a decreasing share of the savings as their rate of output increases. T h e reason for this is that under the Rowan Plan the worker's bonus is calculated as the percentage represented by time saved divided by time allowed. For example, if a worker receives $ 1.00 per hour and takes seven hours to do a job for which he is allowed ten hours, his earnings are calculated as follows: E = TR +

TR

Where E equals earnings, T equals time taken, R equals rate of pay per hour, and A equals time allowed. E = ($1.00 X 7) + [ 1 0 ~

7

( $ i . o o X 7)]

E = $9.10 In this example the bonus for saving three hours of time is $2.10 or 70 cents per hour. If the job is done in six hours, earnings will be $8.40, and the bonus will be $2.40 or 60 cents per hour. If the job is completed in zero time, the bonus is 100 percent of time taken, or zero. T h e merits of this plan arise from its self-limiting feature. With the Rowan formula for calculating bonus it is impossible for earnings to get too far out of line. Bedaux Point Premium Plan. As originally designed, the Bedaux plan was identical to the Halsey 75-25 sharing plan. Recently the plan has been revised so that the workers receive 100 percent of time saved, and accordingly the present-day Bedaux formula is identical to the 100 percent premium version of the Halsey plan and the standard hours plan. T h e unique features of the Bedaux plan are apparent

14

Nature of Wage Incentives

in the system of controls installed for administering the plan. Each job is reduced to a common denominator consisting of units of time required by an average worker working under normal conditions. These units are called "B's." Each " B " represents an average minute of working time. Usually, sixty "B's" per hour represents standard performance, and expected performance of experienced workers is around eighty " B ' s " per hour.20 Taylor Differential Piece Rate Plan. T h e Taylor Differential Piece Rate Plan differs from conventional piece rates in that the Taylor plan provides for two piece rates on each job. One piece rate is considerably higher than the going time wage for the job, and the other is somewhat below the time wage. The higher rate is paid only when workers attain or exceed a given level of performance. This level of performance is established by systematic time study. It is referred to in the literature as a "high task" standard (that is, it represents the rate of output which may be expected from an average skilled worker properly trained for the job and working at an incentive pace). Generally the "high task" standard is approximately 50 to 60 percent greater than the average rate of output maintained by the workers under a time wage system. T h e time wage level of output is referred to as "low task" output. In establishing two piece rates for each job, Taylor hoped to motivate workers with a significant monetary reward for "high task" production and, at the same time, to discourage inefficient workers from remaining with the company. Merrick Multiple Piece Rate Plan. In actual practice the Taylor Differential Piece Rate Plan did not work well. As soon as workers saw that they could not produce at a 20

Several systems similar to the Bedaux plan are described in the literature. See the Haynes-Manit plan and the Dyer plan as discussed in Lytle, Wage Incentive Methods, pp. 246, S53-54.

N a t u r e of W a g e I n c e n t i v e s

15

" h i g h task" level on a given job, they would slacken their pace. T o overcome this weakness, Merrick added another piece rate to the differential plan. T h e first r a t e — t h e lowest price per piece—applies up to 83 percent of "high task"; from 83 percent to 100 percent of "high task" an intermediate piece rate price applies, and for higher levels of o u t p u t the highest piece rate is paid. Gantt Task and Bonus System. T h e Gantt T a s k and Bonus System was designed to remedy a problem similar to that which the Merrick plan attacked: the complications arising w h e n workers failed to qualify for high piece rate. G a n t t did this by guaranteeing hourly earnings and paying the equivalent of Taylor's high piece rate when workers produced at "high task" or above. In short, the Gantt plan provides a guaranteed hourly rate of pay for all work below the "high task" level. For all output above that level, the worker receives 100 percent of time saved plus a bonus of 20 percent of time saved. In this manner, Gantt attempted to overcome opposition encountered with the Differential Piece Rate Plan and to ease the burden on trainees and new employees. Parkhurst Differential Bonus System. T h e Parkhurst bonus plan was designed to keep bonus or p r e m i u m earnings separate and distinct from the employee's regular wages, and hence to avoid the difficulty of having bonus considered a part of wages instead of a reward for work well done. U n d e r the Parkhurst plan, bonus is a definite scale of dollars and cents established for different levels of efficiency on various classes of work. T h e types of work performed in the plant are separated into a n u m b e r of classes. A specific job's classification is based on such criteria as the responsibility of the worker for machinery, the skill required to perform the job, and the physical effort involved in the work.

16

N a t u r e of Wage Incentives

Bonus is determined by comparing actual output per unit of time with a standard established by time study. T h e time wage paid each employee is entirely separate from his bonus earnings and is even paid in a separate check. In summary, there are numerous formulae for relating output to earnings. T h e unique features of many of these arose from specific operating problems. For example: the Halsey and Rowan formulae were designed to remove the fear of rate cutting; the Taylor Differential Piece Rate Plan was designed to force out the slower and less skilled worker; the Merrick and the Gantt plans attempted to mitigate the harshness of the Taylor Differential Piece Rate; and the Parkhurst attempted to maintain premium earnings as distinct and separate from normal hourly wages and thus to maintain the "bonus" characteristic of extra earnings. From the point of view of this study, the differences between the formulae are relatively unimportant. They deal with minor modifications of the basic idea of wage incentives, and so are only incidental to analysis of the general nature of the impact of wage incentives.

II

The Impact of WAGE INCENTIVES

on Manufacturing Costs A N

E M P I R I C A L

S T U D Y

INTRODUCTION SCOPE AND M E T H O D O F T H E E M P I R I C A L

INVESTIGATION

T h e preceding chapter described, in general terms, the nature of wage incentives. It pointed out that, as a management tool, their primary objective is to aid in obtaining minimum costs. T h e present chapter will analyze empirical evidence of their efficacy in attaining this goal. It will do so by describing the impact of wage incentives as observed in a number of case studies. These case studies cover the use of wage incentives in remunerating rank-and-file workers employed on a tenure basis in manufacturing plants. Included in the study are plants manufacturing asbestos textiles, insulations, women's apparel, farm equipment, steel, metal containers, plumbing supplies, hospital supplies, cosmetics, diesel engines, paint, and ribbon. It should be pointed out that the scope of this study has been limited. T h e use of wage incentives in remunerating salesmen, supervisors, and casual labor has been excluded. Furthermore, no attempt has been made to study the use of group incentive plans. These restrictions arise from the limitations of this writer's resources, and from the fact that the excluded materials pose special problems. A further point that should be mentioned with respect

i8

Impact of Wage Incentives on Costs

to the scope of this study is that the case situations investigated have been restricted to relatively well-managed plants. It is felt that the study of such installations provides the most reliable background for j u d g i n g the impact of wage incentives on costs. T h e twenty-six cases included in this study cover plants employing from 100 to 17,000 rank-and-file workers. Most of the plants employed 400 to 700 workers. Selection of the cases was not governed by the size of the plant. It was based on the availability of information, the nature of the manufacturing processes in the plant, and the attention given to administration of the wage incentive system. T h e data on each case were secured by interviewing. W h e r e possible, this was supplemented by a review of the written records of the company. M o r e than 180 representatives of management, rank-and-file workers, and union leaders were interviewed. From case to case there were variations in the number of persons interviewed and the n u m b e r of follow-up interviews. For example: In one case the writer's information came from six years of experience as a participant-observer. In another case information was gathered by intermittent observations over a two-year period. In the other cases data were gathered through less extended studies. In most of the cases both supervisors and workers were interviewed; however, in a few cases it was possible to contact only supervisors, and in several other cases data were gathered solely through interviews with workers. W i t h few exceptions, information was obtained by friendly and informal interviews. T h e writer was fortunate in having an opportunity to conduct a series of courses on wage and salary administration for the officers of a n u m b e r of local unions. T h e rapport established in this relationship helped greatly in obtaining confidential and frank descriptions of how wage incentives affected the ac-

Impact of Wage Incentives on Costs

19

tivities of rank-and-file workers. Another factor contributing to depth of analysis was the fact that executives from several of the companies included in this investigation were students in graduate courses taught by this writer. Their sincere cooperation was most helpful in providing data that normally would have been difficult to obtain. CHARACTER OF T H E

FINDINGS

T o interpret properly the findings of the empirical part of this study it is first necessary to consider some of the limiting factors. It should be pointed out that only twentysix cases were investigated. This raises questions as to the representativeness of the findings. Although analysis of additional cases would have been helpful, it is felt that the number and types covered are adequate for the purpose of this study. T h e cases cover a variety of industrial plants; further, the case findings were checked against the existing literature, and, most important, they have been tested by analytical reasoning. Thus, the empirical part of this study, when considered in the light of the other aspects of the study, is sufficiently inclusive to serve as a guide to understanding the general impacts of wage incentives. Another type of limitation in the data of this study arises from weaknesses inherent in the interview method. T h e raw data supplied by field interviews consist of descriptions, attitudes, and opinions. These are frequently confusing and contradictory. Often informants fail to differentiate "what is" from "what they believe" and "what they desire." Thus, information gained through field interviews has to be weighed carefully to separate facts from beliefs and value judgments. There are further limitations encountered in studying case situations by the interview method. Most people are ignorant of many aspects of their immediate work environment, and are not fully aware of their own motivations. T h u s the interview material tends

20

Impact of W a g e Incentives o n Costs

to be weakened by the prejudices and ignorance of informants. In the course of this investigation a conscious attempt has been made to mitigate these weaknesses in the interview method. Information has been carefully sifted and weighed. Wherever possible it has been checked against written records. Furthermore, attention has been focused on the activities of those interviewed, rather than on their unsupported opinions and attitudes. PRESENTATION

OF THE

FINDINGS

Not only do the scope and method of this investigation establish limitations on its results, but the nature of the subject matter limits the way in which these results can be presented. Wage incentives exist in a system of complex casual relations, and they are used under a variety of conditions. Consequently, it is doubtful whether conclusions of universal applicability may be drawn. However some generalization is possible, for regardless of the conditions under which wage incentives are used, their principal problems are the same. T h e y involve setting and maintaining standards of performance and an "equitable" wage structure. In dealing with these, certain common principles can be developed. Since this investigation is concerned primarily with the general impact of wage incentives, the materials have been selected to illustrate the more common activities and events associated with their use. In this manner the prevailing effects of wage incentives are brought together in an organized fashion, and logical inference as to their contribution to low-cost production is facilitated. 1 i N o attempt has been made to calculate the impact of wage incentives in dollars and cents. T h i s follows from inherent weaknesses in the use of cost data: Unit cost is seldom clearly definable, and is not always susceptible to precise calculation. Unit costs are affected by many things in addition

Impact of Wage Incentives on Costs

21

For purpose of discussion the materials are organized under the following headings: (1) the effect of wage incentives on the general level of plant wages, (2) the operating problems associated with the use of wage incentives, (3) management activities associated with the administration of wage incentives, and (4) an evaluation of the impact of wage incentives on costs.

THE

EFFECT

GENERAL

OF W A G E

LEVEL

INCENTIVES

OF P L A N T

ON

THE

WAGES

W a g e incentives have two important impacts on wages. O n e involves the structure of earnings within the firm. T h e other concerns the relationship of the general level of plant wages under an incentive system to that f o u n d under a time wage system. T h e present discussion is concerned with the latter of these impacts. 2 Statistical evidence indicates that workers paid u n d e r a wage incentive system generally earn more per h o u r than they would under a time wage system. T w o types of studies support this statement. O n e shows the increase in earnings associated with changing from time wages to a wage incentives system. T h e other compares incentive earnings with time wage earnings on similar jobs. In 1943, the Bureau of L a b o r Statistics of the U.S. Department of L a b o r compared the hourly earnings of time workers with those of incentive workers in identical occupations. T h r e e important industries were covered by this investigation: machinery manufacture, cotton-textile to wage incentives. T h e y may change due to variations in lot size, changes in machinery, changes in management efficiency, etc. In the absence of a means for precisely measuring the impact of wage incentives, this writer has observed their operation and proceeded, on the basis of logical reasoning, to infer their impact. 2 For a consideration of the impact of wage incentives on the internal wage structure of the plant, see pp. 30-32.

22

Impact of Wage Incentives on Costs

manufacture, and primary metals. T h e study revealed

fabrication

of

nonferrous

a definite and substantial margin in favor of the workers paid under incentive plans. . . . In roughly half of the occupations in which comparisons were made, incentive workers were earning, per hour, between 10 and 20 per cent more than the time workers. Differences of less than 5 per cent or more than 30 per cent were but rarely encountered and appeared, in most cases, to reflect deficiencies in the statistical data available for analysis.3 T h e general nature of these findings is supported by a later study conducted d u r i n g 1945 and 1946. T h i s study is based on a survey of 34,000 m a n u f a c t u r i n g establishments with about five and one-half million workers, and 21,000 nonmanufacturing establishments with a b o u t one and one-half million workers. It concludes: Generally, incentive workers receive higher earnings than do time workers in comparable jobs, although the size of differential is not consistent from industry to industry. T h e earnings advantage of incentive workers ranged from less than 5 per cent to at least 40 per cent in the individual manufacturing industries studied in 1945-46; in many of the industries the difference was between 15 and 25 per cent. Among the four major manufacturing groups . . . the largest differential appeared in the apparel industries where incentive workers earned from a fifth to two-fifths more than time workers. In the metal working industries, incentive workers most commonly received from a fourth to a fifth more than time workers, whereas in the textile industries the differentials were typically between a sixth and a tenth. T h e chemical industries, in which incentive pay is relatively unimportant, showed no consistent pattern of differences between time and incentive earnings, although in several of these industries the 3 United States Department of Labor, Bureau of Labor Statistics, Effect of Incentive Payments on Hourly Earnings, Bulletin No. 742 (Washington, U.S. Government Printing Office, 1943), p. 1.

I m p a c t of W a g e

Incentives on

Costs

23

difference was small. Among the nonmanufacturing industries in which incentive pay was most important—automobile repair shops and clothing and department stores—the differential amounted to about a third.4 Studies of changes in the level of earnings associated with changing from time wages to wage incentives indicate results similar to the above. In 1944, the Management Consultant Division of the U.S. W a r Production Board investigated eighty-six wage incentive installations in the Chicago area. It found that, on the average, workers' earnings increased by 19 percent. 5 A study made in 1948 showed that earnings of average production employees on incentive, expressed as a percent of straight-time base pay, ranged from 102 percent to 1 5 5 percent. T h e average was 1 2 4 . 4 percent. 6 N o statistical breakdown as to the average bonus earnings was attempted for the plants investigated in the course of this study, but the information gained through available records and interviews corresponds closely with the findings

quoted above. 7 F o r example, in Case N o . 2 3 the

• J o s e p h M . Sherman, "Incentive Pay in American Industry, 1945-1946," Monthly Labor Review, L X V (November, 1947), 535-38. T h e U.S. Department of L a b o r , B u r e a u of L a b o r Statistics, has made a series of studies of industry wage structures. In nearly every case, incentive earnings exceed time earnings for workers in the same occupation. For example, see Wage Structure: Radios, 19-17, Series 2, No. 2 (Washington, February, 1948). See also National Industrial Conference B o a r d , Inc., Wage Incentive Practices, N o . 68 in Studies in Personnel Policy (New York, 1945); American Machinist ( J a n u a r y 3, 1946), pp. 9 7 - 1 0 8 ; Management Review, X X X V I I (October, 1948). s United States W a r Production Board, Addresses and Papers on Wage Incentive Plans and L a b o r Management Relationships (Mimeographed, October, 1944), I V , 4. 6 Management Review, X X X V I I I (January, 1949). Similar results are of indicated by the survey made by the Dartnell Corporation, Experience 117 Companies with Wage Incentive Plans (Chicago, 1948), pp. 5 - 7 . t As used in this study, " b o n u s " refers to earnings in excess of guaranteed hourly rates of pay. For example, if a worker who is guaranteed $1.00 an hour averages $ 1 . 5 0 under the wage incentive system, the extra 50 cents is his bonus. Percentage bonus earned is determined by dividing the bonus by the guaranteed hourly rate of pay.

24

Impact of Wage Incentives on Costs

approximate average bonus was 26 percent; in Case 17, it was 33 percent; in Case 7, it was 30 percent; in Cases 2 and 8, it was 40 percent; and in Cases 1 , 9 , 15, and 26 it ranged from 40 percent to 50 percent. It should be noted, however, that these percentages are not constant. From time to time they fluctuate quite widely. Furthermore, within a given plant the range of bonus earnings is significant. In most of the cases studied, the incentive bonuses on different jobs within a plant varied from zero to 150 percent. O P E R A T I N G P R O B L E M S ASSOCIATED WITH T H E OF W A G E

USE

INCENTIVES

T h e actual operation of wage incentive systems involves numerous problems. 8 T h e nature of these problems changes little from case to case. In fact, analysis of the field studies of this investigation shows that the differences between the problems encountered in the various cases are essentially differences of degree rather than kind. In the discussion that follows these common wage incentive problems are described under two classifications: problems involved in setting standards of output, and problems concerned with the maintenance of an acceptable wage structure. 9 « T h e problems associated with the use of wage incentives have been mentioned by many writers. See David F. Schloss, Methods of Industrial Remuneration (3d ed.; Oxford, England, Williams and Norgate, 1898), pp. 60-86; W. D. Stearns, "Wage Payment Systems in Machine Shops," Machinery, X X V (August, 1919), 1 1 1 5 - 1 6 ; Charles B. Going, Principles of Industrial Engineering (New York, McGraw-Hill Book Co., Inc., 1 9 1 1 ) , pp. 120-25; W. D. Forbes, "Why Piece-Work Is Not Satisfactory," American Machinist, L I I (March 18, 1920), 612; Harrington Emerson, " M y Objections to the Piece-Rate Method of Wage Payment," Industrial Management, L V I I (June, 1919), 470-72; National Industrial Conference Board, Inc., Some Problems in Wage Incentive Administration, and Wage Incentive Practices, No. 19 and No. 68 in Studies in Personnel Policy (New York, 1940 and 1945); Experience of 123 Companies with Wage Incentive Plans, Section I (Chicago, Dartnell Corporation, 1944). 9 It should be pointed out that these classifications are not totally independent of one another. Maintaining an acceptable wage structure is

Impact of Wage Incentives on Costs SETTING

STANDARDS OF

25

OUTPUT

One of the distinctive features of all wage incentives is the fact that an explicit standard of output is determined for each job or task. T h e study of individual cases reveals that in all local plant situations this process involves a fundamental struggle by workers to obtain loose standards of output. 10 T h e workers recognize that their self-interest is served by maintaining liberal standards and by producing at a rate in excess of the standard. Workers employ a number of techniques to obtain loose standards. One of the most dramatic is the deception of management representatives as to the nature of jobs and the general conditions affecting production. Workers' ingenuity in deceiving time study men suggests that numerous loose standards are used in industry. Certainly the evidence uncovered in this investigation supports Henry Dennison's statement that ". . . no time study man living is clever enough to best a moderately clever mechanic and discover the true time." 11 Case No. 6. An illustration of the imagination and inventive faculty of workers was revealed in an interview with a skilled machine operator. This man operated a grinding machine used in the manufacture of roller bearings. He explained that he "put one over on the time study department" by machining his own sample bearing. The sample is the bearing sent down by the engineering department for checking the accuracy of gauges. By manufacturing a sample of smaller dimensions, the worker was able to get increased time for the performance of the operation. All of the scrap manufactured during the period in which the time study was being taken was either smuggled out of the plant or hidden between the walls of the factory. in part a function of setting standards of output. T h e y are, however, sufficiently distinct to be an aid to the organization of the materials. A loose or liberal standard is one the worker can meet without und u e effort; a tight standard is one the worker finds difficult to attain. 11 Henry Dennison, " W h a t Can Employers Do about It?" in Stanley Mathewson, Restriction of Output among Unorganized Workers (New York, Viking Press, 1931), p. 188.

26

Impact of W a g e Incentives on Costs

T h e same worker reported other deceptions, such as using grinding wheels of small outside diameter and bleeding the hydraulic system on automatic machines. 12 Case No. A similar example of worker cleverness in outsmarting time study men showed up in a basic steel plant. A former operator on the skin mill recalled how he had succeeded in obtaining a standard of 120 coils per hour for a job on which the workers could easily produce 300 coils per hour. He did this by distorting the tension on the steel. When the time study men requested that the mill be speeded up, the uneven tension on the steel would cause a breakdown. Most of the time study men interviewed recognized the fact that workers were constantly attempting to obtain loose standards. Case No. 1. T h e time study men in an asbestos textile plant reported a variety of attempts by workers to influence standard setting. These included oiling belts on machines to decrease friction and thus to reduce actual machine speeds; saving up defective materials to be used during time studies; using slow drive gears; and surreptitiously cutting yarn to increase down time. Case No. 7. T h e supervisor of time study in a plant manufacturing hospital supplies and industrial tapes pointed out that because workers attempt to deceive time study men by introducing false motions and avoidable delays his men frequently rate work methods. T h a t is, they arbitrarily reduce the time allowed for a job in order to discount superfluous motions. Case No. 4. One of the time study men in a basic steel plant stated: "Whenever I take a time study I am on my guard. I always expect the workers to try to put something over. For example, on the Number 3 slitter, the one that cuts heavy 12 T h r o u g h o u t this report the material gathered in the course of the field study will be presented in this manner. In accordance with the guarantee of anonymity given to the cooperating companies, their names will not be revealed; however, each case situation is assigned a number so that the reader may recognize illustrations descriptive of a single plant.

Impact of W a g e Incentives on Costs

27

cartridge steel, the workers stalled so much that I refused to set a rate. On the Number 3 bloomer pit the crane operators introduced all sorts of delays. T h e most noticeable one was the taking of six to seven minutes to cool their tongs. Normally they take one half minute for this operation. In my study of the Hallden Shear the operator induced long breakdowns by failing to shut down as soon as the material began to cobble." 1 3 T h e slowdown is a second technique used by workers in attempting to obtain loose standards. It involves concerted action by workers to hold production at a level considerably below their potential output. As such, it is both an attempt to deceive time study men and a collective action to coerce management into liberalizing rates. T h e relationship between the slowdown and the deception of rate setters is revealed by the fact that frequently the mere suggestion that a time study may be taken will slow up an entire operation. 1 4 Case No. 5. This phenomenon is illustrated by the fact that not only did the workers slow down every time a new garment was put into production, but in one case, when a style number was changed without a corresponding change in the product, production immediately dropped. T h e existence of a slowdown is usually inferred f r o m changes in productivity. Sudden and severe drops in output associated with the installation of new standards provides prima-facie evidence of a slowdown. F o r example: Case No. 2. In a steel fabricating plant the workers protested newly established rates by holding their production to 8,582 13 Although anecdotal material is presented in quotation form, is should be noted that the statements are only a p p r o x i m a t i o n s of the wording used. Every attempt has been made to report f a i t h f u l l y the materials gathered in field interviews. However, since most of the data was gathered in friendly, informal conversations, verbatim notes were seldom taken. 11 T h i s type of slowdown is also reported in Mathewson, Restriction of Output.

28

Impact of W a g e Incentives on Costs

pieces per eight-hour shift. Normal production was 14,550 pieces per eight-hour shift. At a later date, in protesting the same standard, output dropped to 5,581 pieces per eight-hour shift. A l t h o u g h it is relatively easy to recognize a slowdown, proving its existence is extremely difficult. Generally the workers rationalize their drop in production in a manner that leaves room for doubt as to the real reason for the decline in output. Case No. 4. T h e problem of conclusively proving the existence of a slowdown is illustrated by the experience of a basic steel company. It had installed a new pickling line. While incentive rates were being determined, the workers were paid their previous average hourly earnings. T h e average output during the period was 279 tons per line turn. At this time 15 men operated the line. From September, 1951, until August, 1952, management applied a new standard to the operation. T h e workers contested the rate, and during this period the average production per line turn dropped to 179 tons. In an attempt to overcome this slowdown, management at one time assigned a supervisor to each of the workers on the line. Despite this, there was no significant change in productivity. It is mainly through hindsight that the magnitude of a slowdown is indicated. Case No. 4. In the pickling line dispute cited above, an arbitration finally established the correctness of management's standard. Soon production jumped from 179 tons per line turn with a crew of fifteen men per turn, to 625 tons per line turn with a crew of nine men per turn. T h e economic loss that workers will incur to sustain a slowdown varies from situation to situation. Frequently it is of considerable magnitude. Case No. One of the costly slowdowns noted during this study was observed in a basic steel mill. The tin plate bundlers

Impact of W a g e Incentives on Costs

29

were put on a wage incentive plan in the latter part of 1950. At that time the workers were averaging 16 to 17 bundles per day. T i m e studies revealed that a more realistic output was 66 bundles per day. From the date of the installation of the incentive plan until the time of these observations (March, 1953) the production was pegged at 16 to 17 bundles per day. Because of this the workers have been averaging around $ 1 , 6 1 5 per hour, rather than the $2.88 they can earn by producing 66 bundles per day. Another device used by workers to coerce management into setting loose rates is the strike. Most frequently the " q u i c k i e " walkout is used. It generally involves a sudden shutdown of a department or section in an effort to force management to make concessions on standards. F o r example: Case No. 1. T h e workers in the Twisting Department were not satisfied with management's disposition of their complaint over a standard. In protest they shut down their machines and marched into their foreman's office. They refused to return to their work until given a guarantee that a more liberal standard would be applied. Occasionally these wildcat strikes start with a few workers or a department and spread to become general walkouts. Case No. 4. In a basic steel plant a dispute over the standards on a new pickling line (see above) resulted in a sitdown strike. In four days this spread to a strike of the entire mill. Fourteen thousand workers were involved. T h e strike resulted in the loss of 366,512 man hours of work, 65,700 tons of basic pig iron production, and 72,800 tons of ingots. T h e total cost of these slowdowns and strikes can only be vaguely inferred. T h e general evidence revealed by this investigation indicates that their costs can be, and frequently are, substantial.

go

Impact of Wage Incentives on Costs

Case No. 2. A steel fabricating plant kept a ten-month record of the approximate loss of production due to work stoppages and slowdowns. It is estimated that, for the period covered, work stoppages resulted in a loss of 26,530 production line man hours and slowdowns cost 1,364 man hours. T h e following excerpts illustrate the data included in this analysis. On December 14, 1944, the workers on the oil line in the Pail Assembly Department shut down for thirty-three minutes to protest their piece rate. This caused a loss of production of six man hours. On August 10, 1945, the pail assembly line shut down for one hour and six minutes to protest a new rate. On August 13, 1945, the pail assembly line slowed down for eight hours to pressure management into loosening a new rate. During this period production dropped from 14,500 pieces per eight hours to 8,582 pieces. On August 22, 1945 the pail assembly line was shut down for two hours and sixteen minutes to protest a new rate. This involved a loss of 10.1 man hours. In the Barrel Department there were a number of slowdowns. From December 17, 1946, to December 27, 1946, production was held to 1,200 barrels per shift. It gradually increased to 2,200 barrels per shift by March 12, 1947. T h e company maintained that satisfactory performance for this barrel line was 2,400 per shift. MAINTAINING

AN

ACCEPTABLE

WAGE

STRUCTURE

In each m a n u f a c t u r i n g plant there exists a wage structure which, over a period of time, has become generally accepted. Usually it reflects the relative worth of the various jobs in the plant. In addition, it tends to rank jobs according to their places in the plant's social hierarchy. Moreover, it is a commonly held value judgment that a worker's earnings should reflect the recognized or accepted value of his j o b as well as his ability in performing on that job. W a g e incentives inject a dynamic element into the

Impact of W a g e Incentives on Costs

31

plant's wage structure. Earnings may become divorced from the individual's merit and the recognized value of his job, and the resulting wage structure may cease to conform to prevailing concepts of fairness and justice. T h i s type of distortion in wage structures was common to the cases covered by this investigation. 1 5 Case No. 1. An illustration of the distortion that may occur in the wage structure under wage incentives is the experience of a plant manufacturing asbestos products. A comparison of the relative ranking of jobs under a job evaluation plan with the earnings under the incentive plans shows considerable disparity. T h e table below illustrates this fact. It shows that average hourly earnings including bonus varied widely from the wage structure that would conform with the ranking of jobs according to the job evaluation plan. For example, the job carrying the greatest number of job evaluation points (machinist supervisor) actually received less per hour than a semiskilled job (pipe covering finisher).

Job

Title

Machinist Supervisor Insulation Dept., Subforeman Carding Dept., Subforeman Electrician, First Class Weaving Dept., Subforeman Preparing Dept., Subforeman Pipe Covering Dept., Subforeman Block Builder, Pipe Covering Lead Sawyer, Pipe Covering Pipe Covering Builder Cloth Loom Weaver Spinner, First Class

Job Evaluation Point Score 355 33 1 305 3°4 298 281 241 236 223 211 186 »77

Average Straight-time Earnings Including Bonus (In cents) 128 124 128 120 142 144 160 165 184 165 121 132

is Similar findings are reported by Reynolds. He states: "In 1947 and 1948 it was not unusual to find production workers in labor grades 8 or 9 who were taking home more money than maintenance men in grades 3 or 4." Lloyd George Reynolds, The Structure of Labor Markets (New York, Harper and Bros., 1951), p. 196. See also Frederick H. Harbison and Robert Dubin, Patterns 0/ Union-Management Relations (Chicago, Science Research Associates, 1947), pp. 139-40.

Impact of Wage Incentives on Costs

32

Job

Title

Finisher Card Operator Spoolers Hook Machine Operator Cloth Cutter, Packing Dept. Mix Picker Operator, Preparing Twisting Machine Operator, Packing Dept. Pipe Covering Finisher Cop Winder, Weaving Dept. Disintegrator Operator Sweeper, Carding Dept.

Job Evaluation Point Score

Average Straight-time Earnings Including Bonus (In cents)

171 165 162

102 81 182

>59 '55 •55 "55 >45 145

89 156 108 126

125

97

Case No. In a basic steel mill the distortion in earnings associated with the use of wage incentives was brought out by the fact that workers in the same job class had widely varying average hourly earnings. For example, in Job Class 9 a Feeder on No. 26 Finishing averaged $2.651 per hour, while a Ross Carrier on a Ram Tractor averaged $1.821 per hour. Furthermore, workers on lower job classes were exceeding the earnings of those on higher rank jobs. For example, Feeders on No. 23 and No. 24 Finishing, both in Job Class 7, averaged $2.75 per hour. This exceeded the earnings of the Feeders on the 72-inch Tandem, who are in Job Class 9; the Catcher and Sticker on the 54-inch Tandem, who is in Job Class 12; and the Welder Operator on the Coil Pickler, who is in Job Class 13. Variations in earnings that occur under wage incentives give rise to many problems. T h e more common of these are the development of pressures for the rapid establishment of standards on new jobs, the tendency for wage incentives to be expanded to cover jobs that are not reasonably suited to such a method of remuneration, the development of grievances over standards of output, the development of animosity among rank-and-file workers, the establishment of ceilings on output, and the development of fraudulent practices. The development of pressures for the rapid establish-

Impact of W a g e Incentives on Costs

33

merit of standards on new jobs. T h e creation of a new j o b immediately poses the question, " W h a t is the correct standard of output?" Since the answer to this question is directly related to the worker's earnings, the existence of nonrated incentive work induces anxieties a m o n g workers. O n e of the consequences of this may be a slowdown. 1 8 T h e pressure to have the new j o b rated as soon as possible is frequently initiated by management. However, if the workers suffer losses in earnings while working on nonrated work, they, too, may agitate for prompt setting of standards. Case No. 1. In the initial installation of wage incentives in an asbestos fabricating plant the application of pressure for the rapid rating of jobs was dramatically illustrated. When the system was partially installed all of the production departments clamored for coverage under the wage incentive plan. Furthermore, in an effort to speed up the rate of installation of standards and to encourage the setting of loose rates, the workers on nonrated jobs slowed down. Under these conditions the plant manager sent the following letter to the supervisor of industrial engineering: "Let's not have time study bog down on any operation that is going to take a great deal of time to set up rates. T h e important thing is to get the system spread among the different departments as quickly as possible. . . . I think it would be a good thing to get a weekly report from Mr. [the time study engineer] as to what progress he is making. Where you see that he has slowed up for any reason, you can take corrective measures." The tendency for wage incentives to be expanded to cover jobs that are not reasonably suited to such a method of remuneration. Sound wage incentive systems require the 1 6 See pp. 27-29, above. In incentive shops production on nonrated incentive jobs is usually 40 to 50 percent of the normal established by time study. T h i s means that when standards are applied, output more than doubles.

34

Impact of W a g e Incentives on Costs

standardization of work. T h e i r satisfactory operation requires that the units of output be definable with precision, and that the conditions of work be maintained with substantial uniformity over periods of time. T h e variations in earnings occurring under wage incentives brings about pressure f o r management to ignore these requirements. T h e workers on unstandardized nonincentive work point out that their earnings decrease relative to those of workers on wage incentive jobs, and they attempt to have the traditional earnings differentials restored. Generally this is manifested as a demand for the installation of a wage incentive system to cover the nonstandardized work. Case No. 8. In a plant manufacturing hospital supplies the maintenance mechanics argued at length that they should be included under a bonus plan. They pointed out that the production workers were averaging approximately 40 percent bonus. Case No. 2. In a plant manufacturing metal containers the workers in the Steel Storage Department—the department that distributes raw materials in the plant—filed a grievance to have an incentive plan applied to their work. When management delayed its reply, the workers marched en masse into the general manager's office and demanded a favorable answer to their grievance. In response to this pressure, management installed a group incentive plan. In its haste it failed to study job methods and failed to describe job conditions. Hence, future administration of the plan was hindered. At the time of this study the workers in Steel Storage were earning a bonus of close to 200 percent. Case No. 4. In a basic steel plant the switchmen and conductors in the transportation department requested that their work be put under an incentive system. Between 1951 and 1952 they submitted four grievances requesting this. T h e company maintained that the establishment of a wage incentive plan for this type of work violated all principles of sound management. T o support their position they hired a responsible and

Impact of Wage Incentives on Costs

35

reputable firm of consultants to study the feasibility of establishing wage incentives for this work. T h e consultants reported that the jobs failed to meet the standards required by wage incentives. However, a slowdown and a wildcat strike led the company to install a specially designed wage incentive plan for switchmen and conductors. The put.

development

of grievances

over standards

of

out-

W o r k e r s j u d g e d the correctness of standards of out-

put b y m a k i n g comparisons b e t w e e n the ease of e a r n i n g bonus pay on a specific j o b with the ease of e a r n i n g it on o t h e r jobs. Variations in the looseness of standards a n d the related variations in earnings b r i n g a b o u t n u m e r o u s grievances. T h e general tenor of these is that " t h e

standard

on m y j o b should be lowered to allow my earnings to be consistent with m y previous earnings or w i t h earnings of o t h e r workers." I m p l i c i t in these grievances is the theory that a correct standard is o n e w h i c h provides a consistent bonus. Case No. 1. T h i s concept is implied by a grievance filed by the weavers in a plant fabricating asbestos. T h e y asked that rates be changed so that each weaver could make a fair bonus no matter what loom was assigned to him. T h e theory is implied in a slightly different form in a grievance submitted by a worker transferred to a new job. His grievance read: " I worked for months on the press cutting machine without making a bonus. O n the other job he [the foreman] took me off of I was making a bonus of three to four dollars a week." T h e concept was applied on a plant-wide basis in a grievance submitted by the workers of the Insulation Department. T h e y stated: " W e have tried to get on an equal bonus basis with the other departments, but the way jobs are rated and the way changes are made from one job to another we can't make it. W e are willing to produce, but we want to see results on our part so we can get our weekly wages in keeping with the rest of the

36

Impact of Wage Incentives on Costs

shop. As to the rates on most of the jobs in our department, hardly any is within reason. We aren't able to make money in keeping with the other departments." Case No. 7. Another aspect of this criterion for judging standards was applied in a grievance contesting the rate established after a method change had been made. The workers in the Ligature Department had been averaging $1.75 per hour on a packaging operation. Management improved the method used for packaging and set a new standard that was more in line with those on other jobs in the plant. Under the new standard the workers averaged $1.14 per hour. The workers submitted a grievance arguing that their rate had been cut. Their argument implied that the new rate should be set to allow the same amount of earnings as did the old one. Although the workers argue for consistency in earnings, the actual application of this criterion is extremely difficult. Consistency by itself is an ambiguous concept. If it is to be a useful measure of the accuracy of rates, the objects being compared have to be precisely defined. Workers, however, often make only those comparisons that substantiate their arguments. T h e y may compare earnings under a given standard with their previous high in earnings, with the high of other workers, or with any other figures that seem to justify loose rates. T h e failure to establish a uniform basis of comparison creates confusion. Furthermore, even if uniform comparisons were made, it is doubtful that a high degree of consistency could ever be attained. T h e reason for this was succinctly stated by one plant manager when he said: " T h e Good Lord made everything different." Practically every aspect of every job is subject to some variation. R a w materials, machines, and workers vary from hour to hour and day to day. Furthermore, it is unlikely that a practical measure for precisely correlating effort and output on all jobs can

Impact of Wage Incentives on Costs

37

ever be devised. Human skills tend to be specific rather than general. Success in one kind of job does not necessarily correlate with success in other jobs. 17 Consequently, it is difficult for an individual to compare standards on different jobs in terms of required effort. A worker transferred from one job to another may experience an inverse relation between his effort and output on that job as compared with his previous job. When variations in job conditions are significant, another type of grievance arises. This involves petitioning supervisors for special allowances to compensate for unusual conditions. In such cases the workers usually argue: "We should not be penalized for conditions beyond our control." Case No. 2. An example of this occurred in a plant manufacturing metal containers. The company changed the raw materials used in manufacturing steel barrels. It began using cold roll steel instead of hot roll. This change brought a grievance from the workers requesting a special downtime allowance when cold roll was being run. When management refused to grant this request, the workers on the barrel line initiated a sit-down strike. The development of animosity among rank-and-file luorkers. A contributing factor to friction in the plant is the fact that the structure of earnings under a wage incentive system often violates the workers' concept of justice and fairness. For example, the old-timer who finds that newer men earn more than he is apt to be resentful. T h e existence of this feeling was indicated by one long-service employee in Case No. 9. He asked the interviewer: " H o w would you feel if you found that kids who have been around only a few months earn ten to fifteen dollars a week more than you do?" « F. A. C. Perrin, "An Experimental Study of Motor Ability," of Experimental Psychology, IV (19*7), «4-56.

Journal

38

Impact of W a g e Incentives on Costs

T h e resentment against the " u n f a i r " earnings structure may take subtle forms. It may be expressed in terms of general unrest and grumbling about the plant, working conditions, supervision, etc. T h i s type of expression is so elusive that it is difficult to trace the role of wage incentives in its development. Frequently, however, the resentment that is fostered by the wage incentive system is apparent in the relations between workers employed on interrelated work. For example, it was noticed that when the work of one department was dependent upon the work of another department, any failure on the part of the controlling department could lead to interdepartment squabbles. Case No. 1. This type of discord was noticed in the textile operations of a plant fabricating asbestos. The operations in the yarn division were all interrelated. If earnings in a given department dropped off, the tendency was to put the blame on the workmanship in the preceding department. Over the sixyear period during which this case was observed, open arguments occurred between preparing and carding departments, between the spinning and the twisting departments, and between the twisting and the weaving departments. Discord also tends to develop between wage incentive workers and time wage workers. T h i s arises from the fact that the bonus earnings of incentive workers upset the traditional differential existing between their earnings and those of time wage workers. Consequently, time wage workers argue that they also should be paid an incentive bonus. T o pressure management into such action, the time workers seek the support of their fellow workers. If active backing is not obtained, discord between time workers and incentive workers may develop. Case No. 10. T h e threat of such discord was observed at a meeting of the union officers of a valve manufacturer. In this plant approximately 80 percent of the workers were employed

Impact of W a g e Incentives on Costs

39

under a standard hours wage incentive plan. These workers were averaging bonuses of 40 to 50 percent. A steward representing the time workers gave a forceful argument for union solidarity in forcing management to pay bonuses to time workers. When he noticed that most of those present were cold to his suggestion, he stated: "You had better back us in getting a bonus or we'll see to it that you won't be able to earn one either." Occasionally friction among workers develops when the plant as a whole is forced into a costly slowdown or strike to support the demands of a special wage incentive group. Case No. 4. An example of this occurred in a basic steel mill. T h e workers on a new pickling line embarrassed their union and antagonized their fellow workers by producing at a level which proved that a costly strike of the entire plant had been unwarranted. One of the workers expressed his resentment in the following manner: "Those ! We went on strike thinking that management was trying to force a tight rate down their throat. Now we know that it was just a bunch of greedy guys trying to clean up. Look what they're earning now!" A persistent cause of friction in wage incentive departments is the distribution of work. T h e existence of loose and tight rates means that some jobs can be completed more speedily and easily than others. Case No. 5. The tendency for workers to pick only soft jobs was noticed in a sewing department of a plant manufacturing women's apparel. T o improve material handling, an overhead conveyor had been installed. This conveyor transported workin-process around the department. T h e individual operators obtained their work by selecting a batch from the conveyor. Toward the end of the week only the difficult jobs were left on the conveyor. A check study showed that the most difficult jobs rode the conveyor for weeks. Case No. 7. The type of problem engendered by inconsistencies in standards was stated by one of the foremen in a plant

Impact of Wage Incentives on Costs

4° manufacturing hospital supplies and industrial tapes. He stated: "I can't live with an out-of-line rate. No matter what I do the girls accuse me of playing favorites. They take it out on me and the so-called favorites." A common cause for animosity among rank-and-file workers is the "rate-buster" 18 (the worker who pushes his output to a level considerably above the norm for his peers). T h e exact techniques used by fellow workers in dealing with rate-busters varies from situation to situation. It appears to be dependent upon the worker's standing in his work group, his recognized skill and ability, and the fear of rate cuts. For example, in Case No. 22 one oldtimer was simply approached by the union steward and given a pep talk on the need for group solidarity. In Case No. 23 a rate-buster was subjected to all kinds of verbal insults. In Case No. 1 a rate-buster had his work sabotaged and was completely ostracized by the work group. In summary, if wage incentives are effective, they stimulate aggressiveness and a spirit of rivalry, which increases the probability of friction and ill-feelings among rank-andfile workers. 19 The establishment of ceilings on output. One of the per18 For an interesting descriptive report of the disciplining of rate-busters and of their relationship to the work group see: Orvis Collins, Melville Dalton, and Donald Roy, "Restriction of O u t p u t and Social Cleavage in Industry," Applied Anthropology, V , No. 3 (Summer, 1946), pp. 1-14; Melville Dalton, " T h e Industrial Rate-Buster: A Characterization," Applied Anthropology VII, No. 1 (Winter, 1948), 5-18. T h i s tendency was summarized by one union officer when he replied to the question: " W o u l d you rather work under a time wage system or a wage incentive system?" He stated: "I prefer time wages. W a g e incentives cause much ill feeling among workers. Some are never satisfied with maki n g out on a good rate. T h e y soon kill the job. T h i s causes all kinds of greed and jealousy." T h i s tendency was noted also by Benjamin Selekman, w h o expressed it as follows: "Intrinsically, piece rates induce anxiety, if not always competitive aggressions, in contrast to the comparative securities of time rates." See " L i v i n g with Collective Bargaining," Harvard Business Review, X X (1941), 30.

Impact of W a g e Incentives on Costs

41

sistent phenomena observed in the field work of this study was the maintaining of ceilings on output. It was noticed that most workers are participants in a concerted effort to keep bonus earnings below a specified amount. T h e maintenance of ceilings on output resembles the slowdown in that both are forms of restricting output. 20 However, the two differ in purpose. T h e slowdown is a device used to pressure management to set loose standards, whereas ceilings on output are maintained to protect workers from rate cuts and other undesirable management practices. In other words, workers use the slowdown as an offensive weapon and ceilings on output as a defensive device. Management representatives point to consistent bonus earning as the most frequent evidence of ceilings on output. Case No. 8. The industrial relations manager in a plant manufacturing hospital supplies stated: "In most of our departments we have unofficial ceilings on output. For example, the girls on the machine making adhesive bandages have made fairly consistent earnings for the past three months." More direct evidence of ceilings on output comes from interviews with union officers and rank-and-file workers. Generally they report specific limits on bonus earnings. 20 Ceilings on output are part of the broader area covered under the general heading of "Restriction of Output." For discussions of this see Mathewson, Restriction of Output. Mathewson found that: (a) restriction of output arose primarily from fear of cuts in incentive rates and of unemployment, and (b) restriction was an important problem among unorganized plants. See also: Burleigh Gardner and David Moore, Human Relations in Industry (rev. ed.; Chicago, Richard D. Irwin, Inc., 1950), pp. 180-92; F. J . Roethlisberger and W. J . Dickson, Management and the Worker (Cambridge, Mass., Harvard University Press, 1939), Chap. X V I I I ; Collins, Dalton, and Roy, "Restriction of Output," Applied Anthropology, V, No. 3 (Summer, 1946), 1 - 1 4 ; Van Dusen Kennedy, Union Policy and Incentive Wage Methods (New York, Columbia University Press, 1945), Chap. IV.

42

Impact of W a g e Incentives on Costs

Case Xo. 23. In a plant manufacturing diesel engines a radial drill press operator stated that he was expected to make exactly 26 percent bonus. Case Xo. 10. T h e local union president of a plant manufacturing valves said: "VVe try to keep the boys in line. We don't want anyone making more than 50 percent bonus." Case Xo. 7. T h e union business agent for a plant manufacturing hospital supplies and industrial tapes revealed that "in this plant the group has set 30 percent as a ceiling on earnings. A few workers exceed this level; but if they get too far out of line they are properly taken care of." T h e manner in which ceilings on output are established is complex. Interviews with those closely associated with setting them reveals that ceilings are determined by trialand-error adjustments to a variety of forces. Furthermore, ceilings are not constant over time, nor are they always u n i f o r m from one department to another in a given plant. Case No. /. In this plant each department had its own ceiling on bonus earnings. At one time the ceiling in the weave shop was 52 percent, in the insulation department it was 60 percent, and in the pipe covering department it was 150 percent. These ceilings only applied to work paid at straight-time rates. For overtime (work paid at time and one half) higher ceilings were recognized. Case No. 10. T h e workers in this valve manufacturing plant reported that on overtime work they raise the ceiling on bonus from 50 percent to 180 percent. However, they pointed out that this was so only for very short-run jobs. T h e determination of the level of ceilings on output is related to the workers' long-run interests. T h e workers fear that unrestricted production will lead to rate cuts and reduced earnings. T h e y also fear that high rates of o u t p u t will become the standard for judging all workers, which may result in pressure being applied to the less productive

Impact of W a g e Incentives on Costs

43

worker, and may establish a basis for tighter standards on f u t u r e jobs. Finally, the workers fear that unrestricted output will reduce employment. T h e y see ceilings as a means of insuring a relatively equitable distribution of a limited amount of available work. T h e maintenance of ceilings on output is a difficult task. Ceilings, in themselves, pose a paradoxical situation in which the worker's self-interest in m a x i m u m earnings is in conflict with the over-all interests of his work group. Often considerable animosity develops among workers over the question of maintaining the group standard of acceptable production. In actual practice the rigidity of ceilings on output is, in part, dependent on the ease with which the work group or its representatives may check the output records of individual workers. Case No. 1. This point was illustrated by the events that occurred in the weave shop of a plant fabricating asbestos products. T h e workers had been accustomed to placing their daily production reports in a cigar box on the foreman's desk. T h e reports were then sent to the front office to be sorted by clock number and to have earnings calculated. One of the clerks in the front office suggested that a special rack be placed in the weave shop to allow workers to file their production reports by clock number. T h e approximate savings to be gained from the use of such a rack was close to two hours per week. Once the rack was installed, the average bonus earned in the department dropped. An investigation revealed that the special rack had made it possible for the work group to police the bonus earnings of all of the workers. Many of the operators had been surpassing the established ceiling of 52 percent. Although evidence of ceilings on output was f o u n d in every case covered by this investigation, it is impossible to say exactly how much production is lost through this practice. Norms f o r judging whether or not a worker is producing at reasonable efficiency are difficult to determine.

44

Impact of Wage Incentives on Costs

Without established norms precise measurement is not feasible. Thus, evidence that workers are maintaining ceilings on output is not, in itself, conclusive proof of worker inefficiency. Many workers may be putting forth their maximum effort to attain the pegged level of production. Consequently, it may be that only the exceptionally fast workers are actually engaging in restrictive practices. Despite these difficulties in measuring the cost of ceilings on output, the evidence gathered in the course of this study indicates that, at times, significant losses are incurred. T h e following episodes may be considered as clues to the loss of production resulting from pegged output. Case No. 3. In a woman's apparel plant it was discovered that many of the workers were leaving the plant early. A study of the practice revealed that at one time one third of the machine operators were working short hours. Many of them came in at 10:00 A.M. and went home at 3:00 P.M. Case No. 10. At a union meeting of employees of a plant manufacturing valves the workers were debating the best argument they could present to management to justify extending the incentive system to the indirect workers. One of the workers suggested that they emphasize that an incentive for indirect workers would bring about greater cooperation between direct and indirect workers, and, hence, would increase production. This suggestion wa5 laughed down. It was pointed out that present incentive workers would not produce more even if the indirect workers were on an incentive. Most of the men were already making their 50 percent bonus and were having difficulty covering up their idle time. A check of nine incentive workers in the meeting revealed that for the previous work day each had stalled from one and one-half to three hours. Case No. p. In a farm equipment manufacturing plant one of the union shop committeemen implied that many workers were idle several hours a day. He stated that his greatest problem was to get men to stay on their jobs after they had made out for the day. "Making out" was defined as earning a 50 percent bonus.

Impact of Wage Incentives on Costs

45

Further evidence of the cost of pegged production is seen in the worker practice of " b a n k i n g " output. T h i s practice involves the worker or work group's b u i l d i n g u p a kitty of processed parts or work tickets which may be turned in if and when needed. 2 1 Case No. 5. In this plant the girls maintained ceilings on earnings by banking their production tickets. They would not report production above the established ceiling. T h e excess would be saved and turned in when it would not indicate excessive earnings. One of the women in this plant was promoted to a supervisory position. She automatically went off the incentive plan and consequently could not get credit for her kitty of $80 worth of tickets. Case No. 23. One of the workers in a diesel engine plant frequently turned in tickets on engines after they had been installed in South America. Case No. 11. A railroad car manufacturing company went off of piece rate. One worker reported that he was caught with a kitty of $800 worth of parts. T w o years after abandoning the incentive system the management of this plant was still running across caches of processed parts hidden on the premises. The development of fraudulent practices. T h e use of wage incentives provides workers and lower level supervision with an opportunity to use sharp practices to further their o w n special interests. Several such practices have been described above. Other activities of this kind center around maintaining a constant level of earnings and/or increasing earnings without related increases in output. Case No. 26. T h e falsification of production records to keep earnings from becoming exceptional was noted in a paint manufacturing plant. T h e industrial engineering department noticed discrepancies between the reported production and material 2 1 For examples of the "kitty system" and a discussion of ceilings on o u t p u t see Don Roy, How, Why and How Much Do Workers Restrict Output? (reproduced by the Committee of Human Relations in Industry, University of Chicago, Publication No. 11, July, 1949), p. 14; Roethlisberger and Dickson, Management and the Worker, Chap. X V I I I .

46

Impact of W a g e Incentives on Costs

transferred by the receiving and stores departments. An investigation of the situation revealed that the foreman screened all work tickets. He reported only that level of production which was consistent with the established ceiling on earnings. Any excess production was held in reserve. It was turned in only if current output failed to provide ceiling earnings. T h e most frequently encountered falsification of records is associated with the allocation of the workers' time. T h e usual practice is for workers to balance out earnings on various jobs so that loose rates do not show in the accounting records. Case No. 23. This is illustrated by a statement of one of the maintenance department workers in a plant manufacturing diesel engines. He pointed out that "the incentive plan for the maintenance department involves a rapid turnover of jobs and the setting of many new rates. Since business declined, the company has been setting tighter rates. During the day the men usually handle several jobs. They push production on the loose jobs and charge some of their time on tightly rated jobs against the loose jobs." In this manner they maintain a fairly high rate of earnings, even though management is setting rates that are more difficult to meet. Occasionally the workers claim downtime allowances in order to attain a desired level of bonus earnings. Case No. 2. An illustration of the practice of padding downtime allowances to increase earnings occurred in the barrel department of this plant. In this department the management agreed that downtime for the barrel line would be calculated on the operation of the double seamer, the fastest machine on the line. T h e workers' technique for increasing earnings was to check out on the double seamer and to keep the rest of the line operating. Then, when a sufficient supply of barrel shells had accumulated to keep the double seamer going at full speed, the workers would check in on it. T h e chief industrial engineer described the situation as follows: "In 1942 the Company agreed

Impact of W a g e Incentives on Costs

47

to let the workers check out for downtime. T h e double seamer was the spot where they checked out. When the double seamer was down the entire line got credit for downtime. T h e capacity of the double seamer was 360 barrels per hour. T h e workers juggled the distribution of their work time so that the double seamer would be shown to run at capacity. At one time the workers claimed to be running the double seamer at 370 barrels per hour. Management called them on this. It pointed out that they had taken excessive downtime and proved this by showing that the maximum capacity of the double seamer was 360 barrels per hour. From that time on we had an average rate of output of 360 shells per hour, and we had forty to sixty checkouts per day. T h e workers continued to claim downtime until the chutes filled up. Then they checked in and ran at capacity." Another type of falsification of records revealed by the field investigation of this study is the misrepresentation of reported production. Case No. 7. In this case it was reported that the accounting department had discovered a discrepancy between the sale of scrap paper and the production reported on labor tickets. T h e quarterly records showed: 1st quarter, 32,575 pounds reported baled and 47,108 pounds sold; 2d quarter, 35,459 pounds reported baled and 49,838 pounds sold; 3d quarter, 42,017 pounds reported baled and 57,942 pounds sold. Investigation of this situation revealed that, because of low bonus earnings on baling of corrugated paper, the worker had charged his time for that work to time wage work. A practice closely related to cheating but more subtle is sacrificing of quality for quantity. T h e tendency for wage incentive workers to increase the quantity of their output while reducing its quality has been noted by many writers. T h e observations made in the course of this investigation tend to substantiate these earlier reports. It was noted that, in situations in which management had not estab-

48

Impact of Wage Incentives on Costs

lished close controls, wage incentive workers tended to sacrifice quality of product, care of equipment, and economy in the use of raw materials for quantity of output. Case No. 23. In a plant manufacturing diesel engines a radial drill press operator stated that he generally ran his machine with speeds and feeds in excess of those specified. He recognized that this increased the cost of cutting tools, but justified his actions on the grounds that it gave him greater freedom in setting his own work pace. Case No. 15. In a plant fabricating pipe covering the workers discovered that they could reduce the number of rejected sections if they increased the density of the product. Since the workers were not penalized for overweight sections, they were not concerned about the increase in material costs and drying time. One foreman summarized the situation by saying: " T h e incentive system puts these guys into business for themselves, and it's to hell with the company. With these sections running overweight it's just like putting a dollar bill in every carton."

MANAGEMENT

ACTIVITIES ASSOCIATED WITH

ADMINISTRATION

OF W A G E

THE

INCENTIVES

Management's activities with respect to wage incentives are directed at avoiding or solving the problems discussed above. For purpose of discussion these may be grouped as: (1) explaining the wage incentive system to employees, (2) setting and maintaining sound standards, (3) installing the standard, (4) avoiding or correcting inequities in the wage structure, and (5) preventing fraudulent practices. E X P L A I N I N G T H E WAGE INCENTIVE SYSTEM TO

EMPLOYEES

T h e first step in installing a wage incentive plan is explaining it to the workers. A common practice reported in this investigation is for management to explain, in a fairly complete manner, the details of its wage incentive plan. It should be noted, however, that this explanation

Impact of W a g e Incentives on Costs

49

is generally limited to the initial installation of the incentive plan. None of the plants studied had a sustained program for explaining wage incentives to rank-and-file workers. A few of the companies reported that they explained their wage incentive plans in the employee's handbook. However, requests for the handbook met the answer: " A t the present time, our handbook is out of print." Interviews revealed that the statements in the employee's handbook are general. T h e following example illustrates their usual form. T h e Management of Plomb Tool Company believes that payment for production work should be based upon several fundamental principles, as follows: 1. It is the policy of the Company to guarantee hourly base rates regardless of how low an employee's production may temporarily fall. 2. Consistent failure to produce up to standard is considered evidence of incapability and, therefore, cause for dismissal. 3. T h e Company shall have sole right to determine what method and equipment shall be used in the performance of an operation. 4. A time standard is set on one method only. When a standard is properly established so long as the standard procedure is followed, the standard is guaranteed against changes regardless of how much an employee earns. 5. Payment of operators shall be in proportion to production. After the standard has been met, a bonus for extra performance is given. A separate fund amounting to one-third of the bonus given is set aside for division among indirect operators for their indirect contribution to the direct operator's production. 6. When the method used on an operation deviates from the standard practice in any way, regardless of who makes the change, a new standard must be set by accepted time study methods.

50

Impact of W a g e Incentives on Costs

7. Suggestions made by employees for changes in method will be paid in accordance with the established suggestion award procedure. 8. When a standard is improperly set in the opinion of either employees or the management and when either desires to adjust the standard in the interest of maintaining equitable differentials in and proper cost of production, the management will attempt to reach an agreement with employees. If an agreement cannot be amicably reached by employees and management, then the machinery for handling grievances will be employed. 22 Frequently information such as the above is included in the labor-management agreement. Case No. 22. One of the most detailed agreements regarding wage incentives was made between a plant manufacturing steel castings and the United Steelworkers of America. T h e agreement provided, in part: "Standard hour incentive rates shall be established by the Company so that the average incentive employee shall be able to earn at least twenty-five per cent above his classified rate of pay. . . . A new or changed incentive standard shall not be subject to any grievances contending that the standard is incorrect until the job has been in production for eighty working hours. During the next eighty working hours while the job is in production, the Union shall have the right to file a grievance. . . . If no grievance is filed, the standard becomes permanent at the end of one hundred and sixty working hours and may not be changed except as provided in Section 4. . . . If a grievance is filed, the standard resulting from the settlement of the grievance becomes permanent. In the event an adjustment is made in a standard as a result of a grievance, the adjustment shall be made retroactive to the installation of the standard. . . . Once a standard becomes permanent, there shall be no change in the standard unless there is a measurable change in the tooling, methods, materials or operations, and 22 National Industrial Conference Board, Inc., Wage Incentive tices, No. 68 in Studies in Personnel Policy, pp. 24-25.

Prac-

Impact of Wage Incentives on Costs

51

in that event, the change in the standard shall only be to the extent of such change in the tooling, methods, materials or operations. . . . All incentive earnings shall be guaranteed daily. . . . T h e classified rates for all jobs shall be guaranteed at all times. . . . An incentive worker or group incentive crew shall receive credit only for work which is of acceptable quality. . . . All downtime, beyond the control of the operator, shall be paid for at the classified rate of pay. . . . N o employee shall be penalized for refusal to work at incentive pace. " I n the event of a dispute as to the correctness of a standard, an earnest effort shall be made to dispose of the dispute in the following manner: (a) T h e Grievance Committeeman in the department where the dispute arose, together with the employee involved and the Foreman and the Company T i m e Study Man shall attempt to settle the dispute, (b) In the event the dispute cannot be settled in Step (a) above, the matter shall then be processed under the grievance procedure. . . . It is specifically understood that a representative of the Union shall be afforded the right at any reasonable time to make a check study of any standard in dispute. " A copy of all methods and time studies shall be given to the Union prior to the installation of any standard. No standard will be installed unless this provision is fulfilled. "Before a time study shall be made, the employees shall be notified that the study is to be made, and the method will be thoroughly explained. "When a regular incentive operator or a regular group incentive crew is required to suspend incentive work for the purpose of making samples, the worker or workers involved shall be paid twenty per cent above their regular base rate or rates for the time spent on such work. ". . . Short order work shall be assigned whenever practical to the junior worker or workers in the applicable classification. For the purposes of the Section, short order work is defined as work performed in connection with the manufacturer of castings where the quantities are insufficient, from an engineering standpoint, to permit establishment of incentive standards." Case No. 7. In contrast with the above a plant manufacturing

5«

Impact of Wage Incentives on Costs

hospital supplies and industrial tapes limited its agreement regarding wage incentives to: "An employee shall be paid for waiting time at his or her time work rate. Waiting time does not include a period of one-tenth of an hour or less for piece workers. . . . Any alleged unfair piecework rate shall be handled as a grievance. In the event an adjustment is found to be justified, such adjustment shall be retroactive to the time the grievance was submitted in writing to the Works Manager. . . . A piecework employee shall be given the minimum guarantee rate for the classification of work in which the piecework job is classified in the event he or she fails to earn as much as the minimum on piecework." Most of the agreements fall somewhere between these two extremes. They define the union's role in setting up or protesting standards and provide the workers with guarantees against rate cutting. 23 2 3 In contrast with the findings of this investigation, a more comprehensive study by the Bureau of National Affairs reveals that: " U n i o n participation in rate setting is basically of three types: (1) T h e union negotiates all rates and its agreement is necessary before any rate may be made effective. (2) T h e union is consulted in advance but the company retains the right to institute the rate even without approval, subject to subsequent union grievance action. (3) T h e company has f u l l control over determination of rates, subject only to grievance action by the union after the rate is p u t into operation. " T h e first practice, joint determination, is prevalent in the apparel industries. Elsewhere, less than 2 percent of incentive provisions call for such procedure. Most often, this f u l l and equal union participation accompanies simple piece rate systems. Arbitration is normally used to resolve disagreements. "Prior discussion with the union, before installation of the rate, is increasingly common. Almost a quarter of incentive clauses embody this requirement. A sizable portion of them, however, are confined to textile and steel industry agreements. Under this procedure, if the union fails to agree on the rate, it is usually on trial for a specified number of days or weeks. A grievance may then be filed and, ordinarily, processed through arbitration if necessary. " B u t the most frequent means of rate setting is still unilateral determination by the company, subject to union complaint. Although only a quarter of incentive provisions spell this out as their procedure, it is apparent that this is ordinarily the means used where the contract does not refer to union participation. Here, too, however, the union may under most contracts dispute the rate u p to arbitration for final adjustment."

Impact of Wage Incentives on Costs

53

Usually they include statements that rates w i l l be guaranteed as l o n g as j o b content or method of operation are unchanged. In addition, they may contain clauses guaranteeing m i n i m u m earnings to incentive workers, regulating the selection of workers to be timed, restricting the use of temporary incentive rates, stating time study procedures, and protecting workers against secret time studies. 2 * A v a i l a b l e evidence indicates that in actuality wage incentives are explained to workers by their supervisors. T o facilitate this, most companies develop a formal or informal program for training supervisors in wage incentive techniques. T h e training of supervisors is frequently carried on in formal classes covering motion and time study as well as wage incentive policies. Such classes may b e taught by outside consultants or by the plant industrial engineer. Formal class sessions usually are supplemented with personal consultations between supervisors and the plant's industrial engineer. Case No. 2. T h e chief industrial engineer in a plant manufacturing metal containers explained that "our foremen have an excellent working knowledge of the wage incentive system. I have personally spent four to five hours in private conferences with each supervisor." Bureau of National Affairs, Inc., Basic Patterns in Collective Bargaining Contracts (Washington, 1949), pp. 15:404-15:405. For other studies of union contract clauses concerning wage incentives, see United States Department of Labor, Bureau of Labor Statistics Bulletins No. 717, Incentive Wage Plans and Collective Bargaining, and No. 686, Agreement Provisions (Washington, U.S. Government Printing Office, 194s). 21 For samples of clauses covering these and related areas, see United States Department of Labor, Bureau of Labor Statistics, Incentive Wage Provisions (Preliminary Draft, April, 1947), and Time Studies and Standards of Production (Preliminary Draft, June, 1947). T h e s e studies contain illustrative clauses selected from the i5,ooo-odd collective bargaining contracts on file in the Industrial Relations Branch of the United States Department of Labor.

54

Impact of W a g e Incentives on Costs

T h i s same company provides its foremen with a detailed manual on wage incentives. T h e following is taken directly from this foreman's manual. It is quoted in considerable detail to illustrate the formal procedures developed for dealing with common wage incentive problems. " T h e wage incentive plan in use in the factory is a type of Manchester plan, that is, a straight piecework system w i t h a daily guarantee. Piecework prices are established by means of time studies, and are guaranteed against any change, except when methods, tools, equipment or layout are altered. " T h e primary responsibility for reporting any change in methods, tools or equipment rests with the Foreman. H e is assisted in this function of his job by the T i m e k e e p i n g , Engineering and T i m e Study departments, but since the Foreman is closer to the jobs being run than the Engineering and T i m e Study personnel, and generally has greater technical knowledge than the timekeeper, he will be most likely to k n o w of any changes before any one else. "Experience has shown that the difficulty of m a k i n g adjustments in incentive rates increases in direct proportion to the length of time that elapses between the change and the rate adjustment. " W h e n any changes do occur, or when operators request a check study, the Foreman notifies the T i m e Study Department or his production supervisor that he wants to have a time study made. T h i s is done by filling out a 'Request for a T i m e Study' form, available in the T i m e Study Office or in the Production Supervisor's office. O n this form he will give the j o b to be studied, the reason for the request, and the date and time of day the job is to be run. After signing the form he gets it approved by the Factory Superintendent or the Assistant General Production Supervisor and gives it to the T i m e Study Department. H e then notifies the department steward a n d the people whose job is to be studied that a T i m e Study Observer has been called to look at the operation. A t this time h e will inform the people that if a piecework price can be developed from the study, they will have the price applied from the start of the job, otherwise they will be paid average earnings.

I m p a c t of W a g e Incentives on Costs

55

" T h e T i m e Study Observer makes an elemental b r e a k d o w n of the j o b by d i v i d i n g u p the j o b on the time study f o r m into a series of short, distinct operations that are practical to time. T h e recorded times are leveled to a normal time by an application of a percentage figure (100 percent b e i n g normal), dep e n d i n g on the performance the operator exhibited w h i l e the study was b e i n g made. T h e leveled time has allowances added to it for relief, fatigue, and miscellaneous delays and the time is then converted to a piecework price. " W h e n the study has been developed into a piecework price, the Foreman and a representative of the T i m e Study Departm e n t g o over all the factors involved to be sure all necessary elements of the j o b are included. W h e n the Foreman is satisfied that the time study covers the job, he signs the study. It is then signed by the General Factory Superintendent or his assistant, and issued for application to the T i m e k e e p e r of the department. " C r e d i t for production on the incentive plan, as well as listing all time worked, by correct labor account numbers usually is the responsibility of the T i m e k e e p e r , b u t he must h a v e the cooperation of the Foreman. T h e T i m e k e e p e r depends on the F o r e m a n to instruct the operators in the correct w o r k methods to follow, the proper way to check in and o u t o n a j o b a n d to answer any questions regarding the application of the incentive plan. T h e T i m e k e e p e r , in turn, must n o t i f y the F o r e m a n of any irregularities in j o b conditions, such as variations in methods, tools, workplace, layout or the n u m b e r of p e o p l e used on the job. AVERAGE

EARNINGS

" T h e process by which the Payroll Department develops an average earning rate is explained elsewhere in this Policy M a n u a l . It is the purpose of this section to e x p l a i n the theory and conditions under w h i c h p a y m e n t of average earnings is made. " A v e r a g e earnings is a form of M a i n t e n a n c e of Earnings. T h e thought behind the payment of average earnings is that an employee is entitled to the rate of pay that he regularly earns o n piecework, w h e n he is assigned to a j o b that w o u l d normally

56

Impact of Wage Incentives on Costs

be classified as a piecework operation but which has had no piecework price established. This does not mean that payment of average earnings is indicated simply because a price has not been established when the employee starts the job. T h e procedure set up by our labor contract is to make a piecework price retroactive to the start of the job after the price has been developed and approved. Deviation from this contract provision is made for the following reasons: " 1 . Because the quantity run is small or other work prevents the time study observer from seeing the job, the run is completed before a study can be made. "2. Mechanical or process troubles prevent application of a standard method of operation. "3. More than one method is used to do the operation due to application of methods or improvements during the course of the job. "4. It is definitely known that the job as it is run at present will have changes in method, tools, equipment, materials or workplace layout within a reasonable length of time. "Because an average earning rate is always greater than a basic daywork pay rate, we have a right to expect a good day's work from any employee compensated in this way. T o o often, we receive a sub-standard performance on such jobs unless the Foreman explained to the people that he will expect a conscientious effort in return for the extra money they receive, and then checked performances on average earnings often enough to demonstrate his interest in the progress of the work. "Before notifying employees that they will be paid average earnings, the Foreman should first check with the T i m e Study Department to be sure no piecework price exists, or no price can be developed for the job as it is being run. He then contacts the General Factory Superintendent or his assistant to get approval for payment on this basis. T o notify the Payroll Department that the operation is authorized for average earnings, the Foreman will make out a request in triplicate on the inter-office communication form, to the General Factory Superintendent, giving the part number, the quantity if the request

Impact of Wage Incentives on Costs

57

is for one run only, and the reason for not using a piecework price. He then has the head of the Time Study and the General Factory Superintendent or their assistant initial the request. One copy goes to the Payroll Department, one to the Factory Superintendent, and one copy is retained by the Foreman for his file. T h e Authorization remains in effect only as long as the conditions which initiated the request exist, and will apply only to those part numbers listed." SETTING

AND MAINTAINING

SOUND

STANDARDS

One of management's constant problems in incentive plants is the setting and maintaining of sound standards of output. T h e usual procedure for working toward this goal is the establishing of a time study specialist to study jobs systematically. T h e exact position of the time study man in the organization varies from situation to situation. In the cases covered by this investigation the prevailing pattern is for the time study men to be included in a staff department handling industrial engineering responsibilities. T h e history of this special department usually is traceable to the growth and development of the wage incentive system. Furthermore, there is a noticeable tendency f o r the time study men to expand their functions to include production scheduling, quality control, cost accounting, and wage administration. Case No. 5. The time study man originally worked under the plant manager. Later he had his own department. Gradually the duties of that department expanded to include responsibility for methods work, quality control, production scheduling, cost estimating, and sales forecasting. Case No. 1. Originally the time study man reported to the plant manager. Later he was placed under the direction of the master mechanic. Next he was put in the cost department and finally, a standards department was established. T h e standards department was primarily responsible for setting rates. However, in time, it took on the staff responsibility

58

Impact of W a g e Incentives on Costs

for cost accounting, job evaluation, union negotiations, and worker training. T i m e study is the c o m m o n technique used f o r setting standards. Most of the plants studied relied on stopwatch studies similar to those described by F. W . Taylor. 2 5 I n several plants the results of time studies were assembled to p r o v i d e standard data. 20 T h i s was most f r e q u e n t where n u m e r o u s short-run jobs occurred. F o r e x a m p l e : Case No. 3. In a women's apparel plant there were frequent style changes. T o cope with the situation, the industrial engineers developed standard data. In this manner they could set standards on new styles as soon as the specifications were known. T h e most recently popularized techniques of building time standards f r o m basic motion times were not used in the plants visited in this study. It should be pointed out, however, that many of the industrial engineers had taken courses in the use of systems such as Methods T i m e Measu r e m e n t ( M T M ) , Motion T i m e Analysis ( M T A ) and W o r k Factor. T h e reason for not using these techniques was exp l a i n e d by one industrial engineer. H e stated: " I have been trained in the use of synthetic times, b u t I seldom use them for setting; standards. T h e main reason for this is that the workers are suspicious of rates that aren't based on stopwatch time studies." INSTALLING

THE

STANDARD

Once a tentative standard has been set, it remains to have it approved and installed. F u r t h e r m o r e , provision 25 See p. 81. 26 Standard data are an extension of the application of conventional time study. T h i s involves the use of elemental time values, determined hy time studies, in setting standards on new jobs. Whenever a new j o b is encountered, it is broken down into elements and the times for these elements are determined from pievious studies. T h i s method increases consistency between standards on related jobs and simplifies the process of setting rates.

Impact of Wage Incentives on Costs

59

should be made for handling grievances over the new rate. T h e procedure for approving new standards varied from plant to plant. In one company the time study man was the only person who approved rates before they became effective. In another company the new rate had to be approved by the proper foreman, the time study man, and the plant manager. In still another several labor relations officers, the accounting department, and the line officials all had to approve a rate before it could be applied. T h e method of informing workers of new standards varied. In several plants the workers learned of new standards as jobs were assigned; in others the management sent a copy of the new standard to the union; and in one plant all new standards were posted on the bulletin board. Case No. 2. An interesting procedure for securing worker acceptance of new standards was observed in a plant manufacturing metal containers. In this plant the responsibility for approving new rates was placed on the foremen. In actual practice the foremen cleared all new rates with the union, so that any objections to the new standard could be discussed on an informal basis. In this manner serious grievances over the acceptability of rates were minimized. Most of the management representatives interviewed reported that if a grievance over standards did arise, it could seldom be settled by arguing the accuracy of the time study. T h e y justified this either by asserting that the workers "don't understand time study," or by pointing out that, since time study is not a science, there is room for considerable disagreement over the correctness of standards. T h e general procedure for settling disputes over rates is to have the workers or their representatives make a cursory time study of the work in question. For example: Case No. 2. T h e general manager of a plant manufacturing metal containers pointed out that the only way in which he could settle grievances over standards was to watch the work

6o

Impact of Wage Incentives on Costs

being performed. He stated: " I don't usually let the time study man participate in our check studies. He is a source of friction and a focal point of worker resentment. Instead, I get the union representative and we both watch the job. T h e foreman is also in on it. He makes certain that the correct method is used. After we watch the job for a while, I ask the union representative to note the rate of output and to compare it with the contested standard. Case No. 5. The chief industrial engineer of a plant manufacturing women's apparel reported: "We settle grievances over rates by having the girls watch the job in operation. If this doesn't end the dispute, we come up to the conference room and I give them a detailed explanation of the standard. Frequently I use motion pictures of the job to prove my point." In several of the cases covered by this investigation the local unions had men trained in industrial engineering techniques. Even so, these unions are careful to avoid direct participation in setting rates. T h e i r principal function is to represent workers in processing grievances. Case No. 22. A plant manufacturing steel castings has two union men designated as time study stewards. These men are trained in time and motion study. Their function is to make check studies of contested rates. On the basis of their studies they advise the grievance committee. The main purpose is to prevent the union from being embarrassed by processing grievances that are obviously unjustified. When the union and management cannot settle a standards problem, they may submit it to arbitration. However, some of the companies covered by this investigation steadfastly refuse to let any third party settle grievances over standards. Case No. 2. In a plant manufacturing metal containers the vice-president in charge of production emphasized that he would never go to arbitration on a grievance over incentive standards. He stated that he did not want any outsider settling

Impact of W a g e Incentives on Costs

61

these cases. T o point out the importance placed on this, he recalled how in 1946 one of the principal issues of the ígi-day strike was whether or not rate cases would go to arbitration. Although most management representatives try to discourage it, the use of arbitration to settle wage incentive grievances is an accepted practice. T h e reason for this was suggested by an international representative of one of the large C.I.O. unions. He frankly admitted that many of the incentive rate cases have political origins. Weak cases are brought up which, for political reasons, cannot be dropped. T h e only way to dispose of these cases is to carry them to arbitration and let the arbiter act as a scapegoat for the unpopular decision. Case No. 4. The need to arbitrate standards grievances is illustrated by the problem encountered in installing rates on a new pickling line. Management and the union struggled to determine the new rate. Because of union politics the case could not be dropped. A sustained slowdown and a costly strike failed to settle the issue. As a final effort, both parties agreed to arbitrate the case. T h e arbitration of wage incentive disputes usually follows the established grievance procedure. Occasionally, however, a special arbitration procedure for settling wage incentive disputes is written into the labor-management agreement. One such clause reads as follows: Case No. 19. "Any dispute as to the equity of a job standard or the evaluation of a particular job not settled as provided above shall not be subject to the arbitration procedure provided in Article X, but shall be settled as follows: "(a) The services of an impartial industrial engineering concern chosen from the list shown below shall be secured. The concern so chosen shall appoint, as a consultant, an engineer experienced and expert in the particular field who shall make an actual study of the disputed job standard or job evaluation. If the dispute involves a job standard, the consultant shall

62

Impact of W a g e Incentives on Costs

determine whether the standard is equitable. If the dispute involves a question of job evaluation, the consultant shall determine whether there has been a change in job conditions since installation of the previous rate, and if so, whether the new rate is based upon proper application of the Company's established job evaluation plan for job rating. "(b) If the consultant finds that a job standard is inequitable or that a job has been evaluated improperly, as hereinabove defined, he shall establish a proper standard or re-evaluate the job as the case may be, in all accordance with the general principles of the Company's established plans for setting job standards and for job evaluation. "(c) Any adjustment agreed upon in Section 5 above or any findings and decisions reached by the consultant shall be final and binding upon both parties for the duration of this Agreement unless further modifications are made in accordance with, and as provided in, this Article." AVOIDING OR C O R R E C T I N G

INEQUITIES

IN T H E

WAGE

STRUCTURE

Management is confronted with a double problem in maintaining an equitable wage structure. It must attempt to maintain consistent differentials between (a) the earnings of workers on incentive pay and those of workers on time pay, and (b) the earnings of workers on different incentive jobs. Failure to maintain the first type of differential leads to pressures to extend the wage incentive system to jobs that are not standardized and for which it is difficult to measure output (such as stockroom work and maintenance department work). In this study three general types of solutions to this problem were observed. One approach used by management to cope with this situation was to give nonincentive workers a bonus based upon the earnings of incentive workers. T h e use of this procedure has been described by Harbison and Dubin in their case study of the Studebaker plant. T h e y report:

Impact of W a g e Incentives on Costs

63

T h e incentive wage system gave rise to a considerable spread between the pay for hourly and incentive workers. . . . Both the union and the company recognized it as a problem. A n answer finally worked out was to put indirect labor on incentive also, with incentive pay tied to the output of the productive workers served. 27 A

second

method

of d e a l i n g

with

this

problem

v o l v e d increasing the base p a y of n o n i n c e n t i v e

in-

workers

a n d l a b e l i n g the increase a b o n u s . Case ATo. 15. In a plant m a n u f a c t u r i n g pipe coverings, the management attempted to correct inequities between maintenance workers and incentive workers by giving the former a flat increase in earnings. T h i s increase, which amounted to 25 percent, was separated from regular hourly pay so as to identify it as a bonus. In actuality, this was not an incentive system. Explicit standards of performance were not set, nor was bonus directly related to output. T h e t h i r d a p p r o a c h to this p r o b l e m i n v o l v e s

putting

t i m e w o r k e r s o n special i n c e n t i v e plans, w h i c h a r e u s u a l l y based o n g r o u p rather t h a n i n d i v i d u a l o u t p u t . Case No. 8. T o provide the maintenance mechanics with an opportunity to earn bonus, a special group incentive plan was installed. Under it the mechanics earned a bonus calculated on the ratio of maintenance hours to the output of the plant. T h e o u t p u t of the plant was calculated in terms of units translated into standard costs as of a given date. Case A'o. 1. In a plant manufacturing asbestos products, the management was faced with the problem of correcting a distortion of the wage structure in the insulation department, in which 90 percent of the work was covered by a Halsey 50-50 premium bonus plan. T h e workers in the department averaged between 40 and 60 percent bonus. In 1946 the department began to get a number of special orders for tailor-made insulations. Since these specials were for 27 Harbison and Dubin, Union-Management

Relations,

p. 139.

64

Impact of W a g e Incentives on Costs

nonstandardized and nonrepetitive work, the company refused to include them under the incentive system. The workers assigned to this special work received a flat bonus of 25 percent. Due to the requirements of the job, only the most skilled workers were assigned to the special orders. These were the workers who had been averaging approximately 60 percent bonus. Several grievances were submitted by these employees. They argued that they were being penalized for being skilled in their work, and that if the less skilled men couldn't be assigned to the special work, the company should at least guarantee the men their average earnings. In response to these grievances, management installed a special group bonus plan. The Rowan formula was used for relating output to bonus. Consequently, the danger of out-ofline earnings was mitigated. The standards used for the special jobs were determined by loose estimates. A written explanation of the bonus plan was presented to the union and the workers appeared to be reasonably satisfied with the system. They applied themselves to their jobs, designed new fixtures and otherwise increased production so that they averaged 50 percent bonus. Management techniques for maintaining equitable earnings differentials between incentive jobs included the following.

Granting

special allowances to maintain worker

earn-

ings. Some managements attempt to avoid extreme drops in workers' bonus earnings by providing that under certain specified conditions the workers will be given their average earnings.

Case No. 5. An example of this occurred in a plant manufacturing women's apparel. The chief industrial engineer pointed out that: "We try to maintain the workers' earnings. We build an allowance of fifteen minutes a day into our standards. In the event that such things as bad materials and machine downtime cause greater delays, we generally pay the girl her average earnings."

Impact of W a g e Incentives on Costs

65

Gradually loosening standards. In some companies management attempts to keep bonus earnings fairly consistent by setting new standards at a level in line with average earnings. Case No. 2. This practice was observed in a plant manufacturing metal containers. The supervisor of industrial engineering reported: "Originally we set standards to allow 20 percent bonus; however, in view of the gradual loosening of standards, we now set them so that a normal worker can earn 40 percent." Helping to maintain ceilings on output. In several plants management consciously or unconsciously contributes to the maintenance of ceilings on output. In several cases this developed from the practice of supplying the union with details on the bonus earnings of the workers. In other cases it grew out of the activities of the foremen and lower level supervisors. Rotating work assignments. Frequently lower level supervisors attempt to maintain a reasonable wage structure by allocating work so as to give each worker an equal opportunity to work on loose and tight standards. Case No. 7. The foreman of an assembly department reported that this was his usual procedure: "The only way I can avoid accusations of favoritism is to rotate the girls so that each gets her share of loose rates." Introducing methods changes. Another device for maintaining a reasonable structure of earnings is the use of methods changes. In most cases management guarantees standards of output only so long as the method of work is unchanged. T h u s , the legitimate means for correcting an out-of-line rate is to change the method. Case No. 1. An illustration of this was revealed in a report by the supervisor of standards to the executive vice-president of the company. It read:

66

Impact of Wage Incentives on Costs

"Bonus inequities in the plant should be corrected. . . . An analysis of the records shows that four departments have average bonuses that are out of line. . . . T h e situation in the first department may be partially corrected by improving the layout of machinery (e.g., a conveyor could be used to carry opened stock to the pickers). T h e installation of such a machine would justify a change in bonus standards. In [the second department] the situation may be corrected by installing an automatic conveyor which will coat, dry and wind six to twelve lengths of tape at one time." It should be pointed out that methods changes are not limited to use as an excuse for tightening standards. I n some cases they were used to loosen standards and thereby to increase earnings. Case No. i. In this plant the twisting department workers complained that their bonus potential was low compared to that of workers in other departments. Rather than lower the standard for the work, management speeded up the machines and thus increased the bonus potential of the workers. PREVENTING FRAUDULENT

PRACTICES

A business agent of the T e x t i l e W o r k e r s of America, C.I.O., stated: "Management is continually thinking of new ways to check production records and to set up a foolproof system. T h e workers spent a lot of time figuring out how to beat these systems. Usually the workers find an easy way o u t . " T h e results of this investigation support this observation. Management usually attempts to prevent workers from cheating. O n e method of doing this is to cross-check actual production with that reported on labor tickets. Case No. i. An example of this occurred in a plant fabricating asbestos. T h e standards department compared materials transferred between departments with production reported on labor tickets. When this check was first instigated, it was discovered

Impact of W a g e Incentives on Costs

67

that, on many jobs with tight rates, labor tickets were never turned in. Instead the workers misrepresented such work in order to get the benefits of looser rates. Another management technique f o r maintaining honest records is to hire timekeepers for each department. Usually the timekeepers report directly to a staff officer. T h e i r function is to distribute labor cards and to see that workers have properly recorded their work time and output. T o simplify m a k i n g production counts, many companies install special counters on their machines. For example, a loom can have a Veeder Root counter mounted on it so that the footage of cloth will be readily determined. Frequently the plants use special mechanical downtime recorders on machines where this factor is important. T h e inadequacy of devices such as these was indicated by this investigation. Case No. 2. T h e company used special charts for recording machine running time. These charts worked on the basis of electrical contacts made when the machine was running. One of the workers revealed that his gang had figured out how to short the recorder and thus show downtime when in actuality the machine was running. Case No. 75. T o record production on looms used for making asbestos blankets, the company installed a meter that registered the number of picks made during the work day. T h e workers beat this system by removing the pulley driving the meter and then turning up the meter by hand. Case No. 1. T o record machine downtime, the company installed Servis Recorders. The Servis Recorder is a simple device consisting of a clock with a long stylus for a hand. When the machine is in operation, the stylus vibrates and scratches a broad band on the waxed face of the clock. The standards department found that these recorders failed to work. T h e reason for this became evident when one worker was observed hitting a recorder with a sledge hammer.

68

Impact of W a g e Incentives o n Costs

AN

EVALUATION OF THE

I N C E N T I V E S ON

I M P A C T OF

WAGE

COSTS

T h e evidence gathered in this empirical study indicates both desirable and undesirable results arising from the use of wage incentives. On the positive side it is evident that: 1. Wage incentives can contribute to increased worker output. In a number of instances rank-and-file workers reported that "when the wage incentive plan was put in we began to help each other and we really increased our output." Furthermore, in certain instances the wage incentive system forced workers to devise improved work methods. Case No. 23. This latter point was illustrated in the case of a plant producing diesel engines. In the testing department the management set such tight standards that the men couldn't make a bonus. After numerous grievances the workers were convinced that management would not lower its rates. Once this conviction was established, the workers cooperated to increase output. By close cooperation and reassignment of job duties they figured a better way of doing the job and thus they were able to earn what they considered an adequate bonus. 2. Wage incentives provide management with a device for circumventing wage stabilization laws. In this they aid in attracting and holding workers. Case No. 26. An example of this use of wage incentives was cited by the chief industrial engineer in a firm manufacturing paints. He pointed out that during the World War II emergency the company installed a wage incentive plan in the drum cleaning department. Loose standards were set to provide a level of earnings that would be helpful in attracting and holding help. 3. Wage incentives allow management considerable flexibility in adjusting to changing market conditions. T h i s

Impact of Wage Incentives on Costs

69

point is closely related to point Number 2 above. T h e reason for separating the two is to emphasize the fact that wage incentives allow downward adjustments in labor costs. Case No. 10. The downward flexibility of labor costs arising under a wage incentive system is illustrated by the experience of a valve manufacturer. In late 195a orders began to fall off and price competition became keen. T o cut costs the company made the workers absorb in their productive time all downtime. Special delay allowances were discontinued. In addition, minor methods changes were made to provide management with an excuse for tightening standards of output. The workers recognized the need for cutting costs, and consequently they were not militant in resisting these changes. T h e tendency for wage incentives to increase the flexibility of labor costs was summed up by a District Representative of the United Steelworkers of America, C.I.O. He pointed out: At the present time [November, 1953] employers in our plants are tightening up on standards. The recession has forced them to take a strong stand on rate setting, and the union isn't in a position to fight this. It isn't like the war days when companies were willing to let up on rates so as to get around wage stabilization. 4. A fourth benefit attributed to wage incentives is the fact that they force management to do a better job of managing. This opinion was stated by the supervisor of masons in a plant manufacturing steel products. He said: The real benefit of our incentive plan isn't the increased effort put forth by the workers. It is the fact that as a result of the installation of wage incentives, we set up proper work methods and determined realistic standards of performance. The wage incentive system has caused us to maintain the best working conditions and to do a better job of scheduling production.

7o

Impact of Wage Incentives on Costs

T h e n e g a t i v e aspects of w a g e incentives h a v e b e e n discussed a b o v e u n d e r the h e a d i n g of o p e r a t i n g

problems.

T h e y i n c l u d e t e n d e n c i e s t o w a r d d e c e p t i o n of t i m e study m e n , s l o w d o w n s , strikes, pressure for r a p i d r a t i n g of inc e n t i v e jobs, pressure to a p p l y the i n c e n t i v e idea to w o r k that is n e i t h e r s t a n d a r d i z e d n o r m e a s u r a b l e in m e a n i n g f u l units, g r i e v a n c e s o v e r standards, ceilings o n o u t p u t , f r a u d u lent practices, a n d f r i c t i o n a n d a n i m o s i t y a m o n g workers. T o these costs of w a g e i n c e n t i v e s should be a d d e d : 1. W a g e i n c e n t i v e s usually r e q u i r e special administrative p e r s o n n e l . I n the cases s t u d i e d in this investigation it was n o t i c e d that the use of w a g e incentives was associated w i t h the h i r i n g of special p e r s o n n e l to h a n d l e t i m e study, timekeeping, inspection, and payroll

calculations.

Case No. 2. T h e need for such extra help is illustrated by the complexity of the payroll calculations in a plant manufacturing steel containers. T h i s company had followed the postwar pattern of wage increases. Because of the piecework system it was forced to put the general wage increases into a separate side-rate for each worker. T h u s , at the time the plant was studied, each worker had (a) an incentive base rate that was guaranteed for all incentive work; (b) an evaluated base rate that the worker received under special conditions; (c) an average hourly earnings rate that was paid under specified conditions, and (d) a side-rate consisting of the general wage increases that had occurred since the war. 2. W a g e i n c e n t i v e s p r o v i d e workers w i t h a d e v i c e f o r d i r e c t l y a p p r o p r i a t i n g part of the benefits of t e c h n o l o g i c a l improvements and for increasing their earnings

without

i n c r e a s i n g t h e skill a n d effort they apply i n their work. 2 8 Case No. 2. T h i s was illustrated in a plant manufacturing metal containers. In the barrel department, management made 2 9 It s h o u l d be p o i n t e d o u t t h a t this aspect of wage i n c e n t i v e s r e l a t e d to their use in m a i n t a i n i n g the flexibility of l a b o r costs. to t h e e x t e n t t h a t w a g e i n c e n t i v e s p r o v i d e a means f o r p u s h i n g a b o v e p r e v a i l i n g m a r k e t levels, they c o n t r i b u t e to excessive costs have a negative i m p a c t .

is closely However, earnings and thus

Impact of W a g e Incentives on Costs

71

a number of technological improvements in the manufacturing process and, at the same time, loosened piece rates by 12.4 percent. These changes provided workers with a 30 percent increase in their bonus earnings. Case No. /. A more detailed account of increasing worker earnings by loosening of standards and improving methods was revealed in a plant fabricating asbestos products. In the spinning department of this plant there had been a gradual increase in average bonus payments. They climbed from 3.2 percent for the week ending January 3, 1942, to 44.1 percent at the end of January, 19.55. An investigation to determine the reason for this significant increase in bonus earnings revealed that raw materials had been improved. A cotton core yarn had been added to the roving; hence, its tensile strength was increased greatly. Consequently, the spinning mules produced more yarn. Fewer ends of roving broke in the spinning process. An added reason for the increase in bonus was the fact that the standard for the most common yarn spun was lowered from 90 meters per hour to 75 meters per hour. This yarn accounted for over 70 percent of the department's production. On the other yarns even greater changes were made. T h e impact of reducing the standard on 10 cut 156 mix yarn from 90 to 75 meters is illustrated by the fact that for the first six months of 1944 the average bonus earned on spinning this yarn was 51 percent. If the old standard of go meters per hour had been used, the average bonus would have been 35 percent. Thus, 16 percent of the increase in bonus was attributed to the loosening of the standard. In summary, this chapter has described the common events and activities observed in the field investigation of this study. It has indicated advantages and disadvantages arising from the use of wage incentives. T h i s empirical evidence, however, fails to provide a basis for generalizing as to the net advantage or disadvantage gained from the use of this device. Before such conclusions may be drawn, it is necessary to reexamine wage incentives in terms of their underlying logic. T h i s task will be undertaken in the following chapter.

Ill The Theory and Logic of WAGE I N C E N T I V E S A

R E E X A M I N A T I O N

T h e preceding chapter has indicated that the use of wage incentives in actual factory situations may both hinder and promote the attainment of minimum-cost production, and has raised questions as to the causes of these economies and diseconomies. T o answer these questions is the task of the present chapter. Accordingly, we shall begin with a consideration of the theory of wage incentives, proceed to an analysis of the logic of their use, and then turn to the implications of this logic under actual operating conditions. In short, this chapter will examine the implications of wage incentives by the process of analytical reasoning in order to find the logical explanation for the patterns of action observed in the case studies. Once these causal factors are determined, it should be possible to indicate the proper place of wage incentives in the management process and the means for attaining maximum benefit from their use. THE

THEORY

OF W A G E

INCENTIVES

Wage incentives may be considered in general or in relation to the environment in which they are used. In general, the theory of wage incentives is an example of the theory of economic motivation. Reduced to simple syllogistic form it holds:

Theory and Logic of Wage Incentives

73

T h a t workers work for monetary reward T h a t for additional remuneration, workers will increase output T h a t , therefore, a system of pay that automatically ties earnings to output increases production T h i s logic is sound. In a "free enterprise" system, monetary reward is one of the basic motivations for work. People obtain the means of satisfying their wants and needs by selling productive power to industrial establishments. 1 Money is the principal medium of exchange and the main standard of value. It is not only an accepted measure of and means of control over things material, but it also has intrinsic value as a symbol of status and security. Furthermore, this logic is consistent with several of the basic moral precepts of our society: (1) that people should be rewarded in accordance with their worth; and (2) that the individual should be "free" to seek his own selfinterest. By rewarding extra effort or skill and by increasing the relative freedom of the worker to better his position, wage incentives conform to these basic values.

THE THE

LOGIC OF THE U S E OF W A G E MANUFACTURING

INCENTIVES

IN

PLANT

T h e use of wage incentives in the manufacturing plant involves not only their general logic, but also the environment in which they operate. Therefore, before analyzing their use, it is necessary to consider the general nature of the manufacturing plant. THE NATURE OF THE MANUFACTURING

PLANT

T h e manufacturing plant as an operating unit is a dynamic system involving coordination of the efforts of in1 A s used here, productive power includes personal services or the use of property.

74

Theory and Logic of Wage Incentives

dividuals performing a wide variety of specialized tasks. Its component parts are numerous. T h e y consist of individual workers, groups, sections, departments, divisions, and so on. Each of these is in itself dynamic. Each may change and create a new state within itself. Furthermore, the various parts of the m a n u f a c t u r i n g plant are mutually related. A change in one aspect or part may cause a change in the relationship of that part to all of the others and a change in the plant as a whole. T h e f u n d a m e n t a l justification for the manufacturing plant's existence is its contribution to the purposes of the enterprise of which it is a part. In the case of private business enterprise, this purpose is the attainment of profit. 2 Profit is attained by maintaining revenues in excess of costs. 3 It is a function of a complex of forces. T h e contribution of the m a n u f a c t u r i n g plant to the attainment of profits is limited primarily to cost minimization. T h e m a n u f a c t u r i n g plant promotes enterprise profits by minimizing the cost of each unit of output. T o achieve this objective, the plant coordinates its component parts and adjusts to technological and economic forces. T h e functions involved in this process are: * i. Determination of the organization of the plant. T h i s 2 Cf. Joel Dean: "A business firm is an organization designed to make profits, and profits are the primary measure of its success." Managerial Economics (New York, Prentice-Hall, Inc., 1951), p. 3. 3 T h e r e are several definitions of profit. T h e one used here is the common business concept of revenue exceeding expense as calculated by conventional accounting methods. * T h e nature of the functions of management has been considered bv a number of writers. See Chester I. Barnard, The Functions of the Executive (Cambridge, Mass., Harvard University Press, 1938), Chap. X V ; E. F. L. Brech, The Nature and Significance of Management (2d ed.; London, Sir Isaac Pitman and Sons, Ltd.. 1948), p. 30; Neil W . Chamberlain, The Union Challenge to Management Control (New York, Harper and Brothers, 1948); and, for an excellent summary of the writings of these and other authors, see Robert T a n n e n b a u m , " T h e Management Concept: A Rational Synthesis," The Journal of Business, X X I I , No. 4 (October, 1949), 225-40.

Theory and Logic of Wage Incentives

75

involves determination of the degree and type of specialization of each of the component parts, and establishment of the relations between each part so that they all contribute to the goal of the plant. 2. Translation of the general goal of the plant into meaningful objectives for each of the component parts. T h e goals of the various specialized units of the plant may not coincide with those of the plant as a whole. Each group or each individual in a group may have specialized purposes or goals; hence, to achieve the over-all purpose of the plant, it is necessary to establish meaningful goals for each unit. These goals must reinforce the attainment of the general objective of the plant. 3. Control of the various component parts of the plant to insure the achievement of their specified tasks. T h i s involves recruiting, selecting, training, and motivating workers, and necessitates the use of budgets, accounting records, and other devices for appraisal of performance. Given this general nature of the manufacturing plant, it remains to inquire into the ways in which wage incentives may contribute to the achievement of its objective. THE

CONTRIBUTION

MANUFACTURING

OF W A G E

PLANT'S

INCENTIVES

TO

THE

OBJECTIVE

Recalling the unique features of wage incentives (that is, that standards of output are explicitly established and earnings are directly and promptly related by a predetermined formula to the relationship between actual output and standard output), it is apparent that a number of benefits may derive from their use. These are: 1. Wage incentives may be a device for attracting labor. T h e y can result in higher earnings, and thus may be a positive force for attracting labor to the plant. 2. Wage incentives may be helpful in placing workers. T h e y provide a means for differentiating between indi-

76

Theory and Logic of Wage Incentives

vidual workers on the basis of output; hence, they may be used as a guide in placing workers where they will produce most effectively. 3. Wage incentives may direct management's attention to training needs. As a device for focusing attention on worker performance, wage incentives may bring to light worker deficiencies that will point the way to a positive program of training. Moreover, the proper administration of wage incentives involves the careful description of the work method associated with each standard. Such job descriptions may serve to guide the workers toward the use of proper work methods. 4. Wage incentives may induce workers to contribute services at an intensity that promotes the attainment of enterprise purposes. This results from the fact that wage incentives integrate the workers' interest in high earnings with the employer's interest in a high level of output. 5. Wage incentives may result in closer management control of work groups and departments. The standards of performance used under wage incentives provide the basis for budgeting labor costs and checking on actual performance. 6. Wage incentives may encourage management to be thorough in determining the best methods for performing work and in standardizing job conditions. This arises from the need for explicit standards of performance on each job. T o be most meaningful, the standard has to be directly related to a given method of performing work; to minimize costs, the standard method should be the most economical. 7. Wage incentives may promote harmonious relations between workers and between the various groups within the plant. By integrating workers' interests, they may help in coordinating the activities of the component parts of the plant. Cooperation may lead to increased output for

Theory and Logic of Wage Incentives

77

all concerned, and, consequently, it may lead to a general increase in earnings. Furthermore, wage incentives may do away with the need for close, driving supervision and thus may provide a basis for more friendly relations between supervisors and workers. If these are the possible benefits that may come from the use of wage incentives, we must next determine the probability of their attainment. T h e proper starting point for such an inquiry is consideration of the assumptions implicit in the application of the general wage incentive theory to the manufacturing plant. APPLICATION

OF THE W A G E INCENTIVE T H E O R Y

MANUFACTURING

TO

PLANTS

In its most common form, the general theory of wage incentives as applied to manufacturing plants maintains: W a g e incentives stimulate the workers to increase their rates of output and, in so doing, bring about reduced costs and increased profits (or decreased losses).5 Implicit in this theory are assumptions concerning the nature of the work, management's ability to set performance standards, the response of rank-and-file factory workers to financial incentives, and the cost structure of the manufacturing plant. ASSUMPTIONS AS TO T H E NATURE O F WORK. For wage incentives to function effectively, a meaningful measure of production must be available. T h i s means that work must be measurable in units that are uniform and distinct. In addition, measurement of output in these units must correlate closely with the skill and effort expended by the worker in maintaining different rates of output. These requirements imply two things: First, work must 6 As stated here, the theory of wage incentives concerns only the impact of wage incentives on worker output and its consequent effect on profits. It does not deal with the other possible effects enumerated above (such as improved selection and training).

78

T h e o r y and L o g i c of W a g e

Incentives

be standardized. Without standardization of the work, differences in measured output may be caused by variations in j o b conditions rather than by differences in the worker's diligence and effort. Second, the worker must have a definite and recognized role in determining his own rate of output. 6 If the worker's efforts do not directly influence output, he lacks freedom to influence his earnings, with the result that wage incentives cannot serve as an inducement to increase production. ASSUMPTIONS AS TO T H E SETTING OF STANDARDS OF

PER-

FORMANCE. O n e of the distinctive features of wage incentives is the fact that standards of performance are explicitly stated for each j o b or task. F o r wage incentives to function effectively, it is necessary that these standards of performance be efficiently determined. T h i s means that the standard must not be so high that it will discourage workers from producing or so low that it will result in excessive labor costs. Furthermore, the cost of setting and maintaining standards must not be excessive. ASSUMPTION

OF

THE

RESPONSE

OF

RANK-AND-FILE

TORY WORKERS TO FINANCIAL INCENTIVES. W a g e

FAC-

incentives

assume that rank-and-file factory workers respond positively to financial incentives, and that this response takes the form of an increased rate of output. ASSUMPTION AS TO T H E COST STRUCTURE OF T H E

MANU-

FACTURING PLANT. Implicit in the theory of wage incentives is the assumption that an increased rate of output lowers unit costs. T h i s implies that the plant incurs certain costs that do not increase proportionately with increases in the rate of worker output, and that the increase in the rate of output resulting from the use of wage incentives therefore brings about a reduction in the marginal cost of each incremental increase in output. « As used here, "rate of output" refers to the quantity or amount of output produced per unit of time.

Theory and Logic of Wage Incentives IMPLICATIONS UNDER

OF

OPERATING

THE

LOGIC

OF

WAGE

79 INCENTIVES

CONDITIONS

G i v e n these assumptions, what are their implications for the use of wage incentives in the m a n u f a c t u r i n g plant? Essentially this is determined by the extent to which they apply to the m a n u f a c t u r i n g situation. Hence, the answer to this question involves consideration of: (1) the extent to which uniform and m e a n i n g f u l measures of production are available; (2) the extent to which standards of performance may be efficiently determined; (3) the extent to which wage incentives motivate workers to increase their rates of output; and (4) the extent to which increases in the rate of worker o u t p u t will decrease marginal costs. THE

DETERMINATION

MEASURES

OF

OF

UNIFORM

AND

MEANINGFUL

PRODUCTION

T h e process of m a n u f a c t u r i n g involves a wide variety of jobs and specialized tasks. For many of these it is possible to measure o u t p u t in units that actually tend to indicate worker efficiency. However, the plant is a dynamic system existing in a dynamic universe. From time to time every aspect of production processes changes. T h e a m o u n t of change may be small or large. T h e important fact is that raw materials, machinery, and workers vary from day to day and from hour to hour; hence, jobs may never be completely standardized. It is always possible that variations in measured o u t p u t may reflect changes in j o b conditions rather than changes in worker skill or effort. Furthermore, in most plants there are jobs in w h i c h the individual has only a limited a m o u n t of freedom in determining his rate of output. T h u s , the possibility of measuring work in units that correlate perfectly with worker skill and effort is remote. T h e best that may be obtained is an approximation; the worst is a complete

8o

Theory and Logic of Wage Incentives

lack of correspondence between the worker's skill and effort and his measured output. T h e implication of the above is that wage incentives should not be applied if work cannot be measured in meaningful units, or if the worker has little or no control over his own output. It follows, then, that in many manufacturing plants there are jobs to w h i c h wage incentive systems should not apply. THE DETERMINATION

O F STANDARDS O F

PERFORMANCE

Writings in the field give the impression that the efficient determination of standards of performance is to be accomplished by the "science" of time and motion study. Hence, at the outset, it is necessary to consider this approach to rate setting. 7 F. W . T a y l o r was the first to popularize the use of time and motion study. H e indicated the importance of such studies in the following manner: Time and motion study are the accurate scientific methods by which the great mass of laws governing the best, easiest and most productive movements of men are investigated. T h e y substitute exact knowledge for prejudiced opinion and force in determining all the conditions of work and pay. . . . As reasonably might we insist on bargaining about the time and place of the rising and the setting sun . . . [as bargain over the results of time and motion study].8 T h i s leads to consideration of the question of whether or not motion and time study actually provide for scientific determination of standards of performance. As used here, "scientific" refers to a procedure agreeing with proven laws and resulting in accurate and exact standards. t In this study the terms "rate of output," "rate," "standard of performance," and "standard" are used synonymously. » Frederick \V. Taylor, quoted in Robert F. Hoxie, Scientific Management and Labor (New York, D. Appleton and Co., 1915), p. 40.

T h e o r y and Logic of Wage Incentives

81

T h e proper approach to this question is to state clearly the nature of time and motion study, and then, to examine their validity and reliability. Properly applied, the procedure for setting standards by time and motion study involves the following steps: 1. Studying the methods used to perform the job 2. Determining of the best method for performing the job 3. Standardizing of job conditions and of the method of doing the work 4. Dividing the job into elemental motions 5. Recording the time required for each element 6. Determining the average time for each element (This involves carefully noting all unusual elemental times and explaining the reasons for variations from the average times.) 7. Adjusting the average observed time for each element to establish the time required by a "normal" worker 8. Determining allowances for personal time, unavoidable delays, and fatigue 9. Adding all of the "normal" elemental times and the necessary allowances to get the standard time for the job T o be scientifically valid, the technique of time study must measure what it purports to measure; that is, the quantity of work which a "normal" worker will produce in a given period of time, or, conversely, the length of time required by a "normal" worker to produce a given quantity of work. Thus, in part, the validity of time study rests on the concept of the "normal" worker. This concept has been described in a variety of ways. Carrol describes the requirements of the normal worker as: The equivalent of effort expected in return for market value of the base rate of pay. It should be the same amount of effort

82

Theory and Logic of Wage Incentives

for all types of work, and the variations in the normal skill required should be compensated for by classified base rates. 9 Barnes defines the normal operator as: T h e operator who . . . would possess that amount of skill that would be expected of a man who knew his job well. He should be able to do accurate work, know how to handle his tools, and work to specifications. He would have to be more than a beginner, although he would not be expected to have that super skill which comes only from many years of experience in a special field, or because he is especially well adapted to the particular work. This average operator would be expected to exhibit an average amount of effort or speed. T h a t pace used by operators working without incentive (that is, paid by the day) with reasonably good supervision is what is meant by average effort. T h e majority of workers should be able to do more than this average.10 T h e point to be emphasized is the vagueness of the criteria used for guiding technicians in the application of time and motion study. T h e " n o r m a l " output of a "norm a l " or "average" worker is a concept existing only in the m i n d of the time study man. It is an abstraction which can be applied only in a vague and general way. T h u s the validity of time study results can never be precisely and definitely established. T i m e study not only lacks a clearly defined goal, it is also of questionable reliability. T h e same technician timing the same j o b at different times will frequently develop conflicting standards. Similarly, several technicians studying a given j o b at the same time often produce varying results. 11 » P h i l Carrol, Jr., Time Study for Cost Control (New York, M c G r a w - H i l l Book Co., Inc., 1943), p. 88. 10 R a l p h Barnes, Motion and Time Study (New York, J o h n Wiley and Sons, 1940), p. 276. 1 1 Barnes indicates that it is not uncommon for two trained observers to come up with 15 percent variations in performance standards (Motion

T h e o r y and L o g i c of W a g e Incentives

83

T h e lack of reliability in the results of time study arises from the assumptions implicit in its use. It is assumed that jobs are standardized; that a trained observer can accurately measure the time taken to perform each element of the job; that the time study observer can calculate the correct average time for each element; that he can adjust this average time to represent the average for a " n o r m a l " worker; that allowances for fatigue and unavoidable delays can be calculated; and that the total time taken to perform a given job can be determined by adding up average "normal" elemental times and allowances for personal time, unavoidable delays, and fatigue. Critical analysis of these assumptions indicates that in actual practice time study is little more than a systematic method of guiding judgment. Jobs are seldom standardized. A study taken at one time may not be representative of job conditions at a future date. Human errors enter into the reading of the stop watch. T h e calculation of a representative sample of observed times involves arbitrary selection of statistical techniques. Leveling (the adjusting of average observed times to represent the time taken by a " n o r m a l " worker), is an attempt to use subjective judgment to attain an indefinite goal. Calculation of fatigue allowances involves measurement of phenomenon which, in itself, cannot be objectively defined. Determination of allowances for unavoidable delays assumes that such delays are predictable and that the observations taken are a basis for making that prediction. 12 and Time Study, p. 277). In the field work of this investigation it was found that standards set by different time study men vary over a wider range. Furthermore, it was noticed that individual time study men tend to develop definite biases in their rate setting. 12 A number of writers have pointed out these weaknesses of the time study technique. See Hoxie, Scientific Management and Labor, pp. 46-47; R a l p h Presgrave, The Dynamics of Time Study i2d ed.; New York, McGrawHill Book Co. Inc., 1945); William Gomberg, A Trade Union Analysis of Time Study (Chicago, Science Research Associates, 1948); R . S. Uhrbrock,

84

T h e o r y and L o g i c of W a g e

Incentives

In summary, analysis of the technique of time study leads to the conclusion that: In face of the lack of conformity in methods, lack of clearly defined or explicitly measurable goals, lack of consistency between the results of various engineers, etc., one sees that standard setting is a matter of refined judgment. Ultimately standards are set on the basis of authority not scientific study.1* In the absence of a science, the determination of standards is achieved by domination, compromise, or integration. Determination of standards by domination occurs when one group—either the workers or their supervisors— forces its concept of the correct standard on the other. Frequently the use of such a technique involves a struggle to determine or to test the power of the respective parties. T h e use of compromise involves both sides making concessions as to what they will accept as a "correct" rate. Compromise is a method of mutualizing dissatisfaction. A t its best it is only a truce, so it tends to break down. T h e process of compromising on the determination of standards implies an underlying power struggle. Compromise represents a settlement associated with a fear of or appreciation of the cost involved in attempting to set standards by domination. Integration of the interests of rank-and-file employees and supervisors involves emphasizing the common interests of both parties. It is doubtful that the integration of interests may carry far in setting standards. Wage incentives "put workers into business for themselves." In so doing they divorce the interests of rank-and-file workers A Psychologist Looks at Wage Incentive Methods (New York, American Management Association, 1935); Adam Abruzzi, Work Measurement (New York, Columbia University Press, 195s). 1» M. Mundel, Systematic Time and Motion Study (New York, PrenticeHall, Inc., 1947), pp. 155-56.

T h e o r y and Logic of W a g e Incentives

85

from management's interests. 14 T h e worker's interest is to obtain standards that allow high earnings and little physical effort. T h e limiting factor in pursuit of this interest is the substitution of capital for labor; that is, the point at which labor costs become so excessive that management eliminates the worker's job. Thus, it appears that setting standards of performance will inevitably be associated with discord. Neither domination nor compromise is a procedure for gaining wholehearted acceptance of standards, and integration of interests tends to have only limited applicability. T h e very mechanics of wage incentives emphasize fundamental differences between the points of view of rank-and-file workers and management. Hence, it follows that the setting of standards will tend to be a center of conflict and that workers may be expected to use deceit, subterfuge, and economic pressure to forward their own special interests. W A G E I N C E N T I V E S AS A D E V I C E F O R

MOTIVATING

W O R K E R S TO I N C R E A S E T H E I R R A T E OF

OUTPUT

Since the motivation of workers by incentive plans is simply a special aspect of the general problem of motivation, it should be helpful to begin with some generalizations. MOTIVATION IN G E N E R A L . Motivation is common to all human activities. It arises from specific needs of individuals. These needs may be biophysical or cultural in origin. Society channels them, provides all sorts of substitutes, and prescribes the socially accepted means of satisfying them. In a general way, the principal motivations operating in our society arise from the following needs: 1. Biological—such as the need for sustenance, shelter, physical well being Management is used here to signify those directly responsible owners.

to

86

Theory and Logic of Wage Incentives

2. Social—such as the need for approval, status, companionship, feeling of belonging 3. Psychological—such as the desire to be useful, the need to excel, the need for personal power, the need for security, for pride in work, for aesthetic or religious satisfaction These classes are not totally independent of one another, but they are sufficiently distinct to be useful in this analysis.15 Although these needs are common to individuals in our society, their intensities and the means for gratifying them vary from person to person. Each individual develops his own personal pattern of feeling and reacting to his environment. T h i s pattern is uniquely determined by the interrelation of his inherited capacities and his previous and present environment. Early in life, he learns to act in a manner prescribed by his environment and develops his own patterns of feeling and thinking. These early patterns form a basis for all subsequent learning and give the individual his peculiar, personal, idiomatic way of thinking, acting, and feeling. T h e y create a set which in turn forms the foundation upon which new experiences are integrated into the individual's personality. He shapes, distorts, and otherwise interprets every fact and event to fit his own framework. This means that the ambitions and needs of each individual tend to be unique. T h e y are to a considerable 1 5 See K i m b a l Y o u n g , An Introduction to Sociology (2d ed.; New York, T h e American Book Company, 1949), pp. 102-4. T h i s classification is not Young's but, in this writer's opinion, is consistent with his concepts. See also L. F. Shaffer, Psychology of Adjustment (Boston, Houghton Mifflin Co., 1936), pp. 100-108. Douglas McGregor lists human needs in three categories closely related to the above: physical (food, shelter, rest, etc.), social (the need for companionship, acceptance, etc.), and egoistic (the need for a sense of achievement, self-respect, etc.). " T h e Supervisor's Job," address before the Management Forum, E. I. du Pont de Nemours and Co., Inc., Wilmington, Delaware, A p r i l 16, 1948.

Theory

and L o g i c

of W a g e

Incentives

87

degree predetermined by previous development and inherited capacities. T h e same activity will mean different things to different individuals, and a given need may be satisfied in a variety of ways by different personalities. WAGE INCENTIVES AS A MOTIVATING DEVICE. T h e

general

logic of wage incentives has already been described.

At

this point it is desirable

in

to enlarge on

the manner

which they may affect worker output, and thus to draw attention to their potential in motivating workers. In our society, money is a m e d i u m of exchange. It is important not only as a means of command over most of the necessities of life, but also because of the distinctions it procures and the status it bestows. 16 T h u s , the financial reward offered under wage incentives may be expected to be a positive force in motivating workers. A second motivational aspect of wage incentives arises from the establishment of an explicit standard of

per-

formance. It is a common-sense observation that the mere act of providing a point of reference for evaluating workers' performances will guide them in setting output goals. 1 9 Frequently discussions of financial incentives overlook the fact that high-paying jobs tend to symbolize increased status. In m a n y plants opportunity for occupational progress is slight and the drive for status is concentrated on getting more money for d o i n g the same job. Reynolds found this to be true in studying the labor market of New H a v e n , C o n n . H e reports: " T o m a n y workers, indeed, more money is virtually the whole m e a n i n g of occupational progress. O u r first interview schedule contained the question: ' H a v e you had any promotions since you started work here?' T h i s question had later to be rephrased, because a large proportion of workers answered it by saying, ' O h yes, I've had three raises since I came with the C o m p a n y . ' " M o r e money may be obtained in a variety of ways. It may come f r o m plant-wide wage increases. It may come from a g r a d u a l loosening up of time standards on an incentive job. . . . "It is wrong to attribute this drive toward higher earnings simply to a desire f o r more money to spend. It is due partly to a c o n t i n u i n g desire for occupational progress, under conditions in which the only possible evidence of progress is higher earnings." L l o y d George Reynolds, The Structure of Labor Markets (New York, Harper and Brothers, 1951), p. 153.

88

T h e o r y and L o g i c of Wage Incentives

T h e establishment of explicit, realistic standards of performance provides a means for stimulating workers to high level output. 17 T h e importance of standards in motivating workers was experimentally studied by Mace. 18 He found that when the standard of performance was fixed, the subjects did not progress so well as when they were told to do their best. Poorest performance was observed when members of a group were told to aim at surpassing their previous week's record. In investigating to determine why the standards of output were not effective in motivating the group, Mace discovered that the subjects' skill had developed to a level at which they could reasonably expect to reach the standard. Further experiments revealed that by using a moving standard, based on the level of the individual's development and on a general knowledge of the rate of learning for the task, the best results were obtained. Thus, standards of performance were most effective where each standard was adjusted to the level of skill and ability of each individual worker. T h e implications of this for the setting of standards under a wage incentive plan are: (1) that the standard may be effective as a pacing device; and (2) that the effectiveness of the standard in bringing forth maximum production will be determined, in part, by the level of effort at which the standard is set. For the more highly skilled or those who work at a more intense pace, the existence of a standard of output which does not challenge their abilities may serve as a brake rather than an accelerator to their production. Mace's study focuses attention on an important aspect of individual differences (their role in determining re17 It should be pointed out that the setting of realistic standards of output may be done independently of wage incentives. 18 C. A. Mace, Incentives: Some Experimental Studies, Industrial Health Research Board, Report No. 7s (Great Britain, 1935).

T h e o r y and Logic of Wage Incentives

89

sponse to a given schedule of output). Another aspect is their importance in determining the response to financial reward. 19 T h e needs of individuals vary widely. One man's feast can be another man's famine. Another implication of the general nature of motivation is that a worker's motives may be in conflict. A worker's desire for social approval can clash with his desire to maximize money income. Recent "human relations" literature has focused attention on this conflict. Mayo typifies this trend. He states: The studies of actual industrial situations which have been carried on show that the desire to stand well with one's fellow, the so-called human instinct of association, easily outweighs merely individual interest and logic of reasoning. . . . The Hawthorne experiment brough out "inconsistencies" of behavior that astonished the engineers and led ultimately to the conclusion that associative instincts overshadowed material conditions as determinants of productivity.20 A further implication derived from the nature of worker motivation is that, in some instances, the workers can view wage incentives as punishment for low output, rather than as a positive inducement to increase output. Hence, to 19 These variations are most noticeable where marked cultural differences exist. See: Allison Davis, " T h e Motivation of the Underprivileged Worker," in William F. Whyte, ed., Industry and Society (New York, McGraw-Hill Book Co., Inc., 1946); Orvis Collins, Melville Dalton, and Donald Roy, "Restriction of Output and Social Cleavage in Industry," Applied Anthropology, V, No. 3 (Summer, 1946), 1 - 1 4 ; Melville Dalton, "Worker Response and Social Background," The Journal of Political Economy, LV, No. 4 (August, 1947), 323-32. An interesting subject for further research is the relation of the personality characteristics of workers to their responses to wage incentives. Some work in this area has been done by J . G. Friend and Ernest A. Haggard. T h e i r study was published for the American Psychological Association under the title Work Adjustment in Relation to Family Background: A Conceptual Basis for Counseling (Stanford, Stanford University Press, 1948). 20 Elton Mayo. This quotation has been taken from the loose-leaf digest accompanying The Social Problems of an Industrial Civilization (Boston, Harvard University Press, 1945). See also W. B. D. Brown, "Incentives within the Factory," Occupational Psychology X I X , No. 2 (April, 1945), 82.

go

T h e o r y and Logic of Wage Incentives

some workers wage incentives can signify a system of penalties rather than one of rewards. The lack of a definite and precise measure of the correctness of standards has serious implications for the motivation of workers. T h e first of these is that workers' desires for financial reward or a relaxed work pace may lead them to use pressure and deception to secure loose rates. If this occurs, wage incentives may not actually lead to increased output or decreased costs. A second inference derived from the problem of setting and maintaining standards is that rates may be of varying looseness; hence, variations in earnings may be more a matter of a worker's luck or trickery than of his increased effort and diligence. 21 If this occurs, the rationality of the wage payment system is weakened and its impact in motivating workers is reduced. A further implication of the problem of setting standa r d s is that the v a r y i n g tightness a n d looseness of Tates

injects a dynamic element into the plant earning structure. Traditional earnings differentials are distorted. T h e plant status quo is upset and the following tendencies develop: 1. Workers refuse to accept promotions to more skilled jobs that do not provide a greater earnings potential. 2. Workers on jobs with low earnings potentials apply pressure for the relaxation of standards. 3. Friction tends to develop between nonincentive workers and incentive workers, and between high-earnings and low-earnings groups. 4. Nonincentive workers bring pressure to bear for the application of wage incentives to their work. These actions are directly related to the psychological and sociological motivations of workers. 21 This tendency has been noted by Norah Davis in "Some Psychological Effects on Women Workers of Payment by the Individual Bonus Method," Occupational Psychology X V I I I , No. * (April, 1944), 53-62.

T h e o r y and L o g i c of W a g e Incentives

91

Finally, the problem of setting and maintaining standards of output enhances workers' fears of rate cuts and thus encourages ceilings on output. In tenure employment, the desire for financial reward is seldom approached from a short-run point of view. T h e workers are interested in m a x i m u m income over a sustained period of time. T h e y recognize that unusual or exceptional earnings will stimulate management to change methods and to apply new standards. Hence, to maximize their long-run income, the workers establish and police ceilings on production. In summary, wage incentives may work in a positive or negative manner. T h e y release forces that may increase worker motivation. However, the paradoxical nature of worker motives and the problems associated with setting and maintaining standards of output tend to offset this response. EFFECT

OF

INCREASED

OUTPUT

ON M A R G I N A L

COST

One of the assumptions implied in the theory of wage incentives is that the increased output stimulated by a system of payment geared to results will bring about decreased unit costs and increased profits. Whether or not these results will be forthcoming depends upon the impact of wage incentives on marginal costs and on the nature of the demand for the firm's product. 22 A case will serve to illustrate this. If it is assumed that a firm produces one item, that the demand for the item is perfectly inelastic (that is, the same quantity will be demanded at any price), that there is no substitution of 22 In this discussion, it is assumed that the wage incentive used rewards workers in proportion to their increase in output. For e x a m p l e : If workers increase output 10 percent over standard, they receive a 10 percent increase in earnings. If the standard of output is above the level of past performance, the wage incentive plan will result in a direct decrease in direct labor costs. (Here it is assumed that workers produce at standard efficiency or better.) If the wage incentive is a sharing plan, such as the R o w a n or the Halsey, management will automatically share in any savings in direct labor cost.

92

T h e o r y a n d L o g i c of W a g e Incentives

labor for capital, and that wage incentives cause the workers to increase their rate of output, then to produce the given output the firm will require fewer workers. With the total output of the firm constant, the possibility of reducing total unit costs by spreading fixed and semifixed costs to a greater number of units is ruled out. T h e only savings derived from the increased rate of output will be those brought about by the reduction in the size of the firm's labor force (for example, such items as vacation pay, holiday pay, pension costs, and health insurance will be reduced). However, offsetting these would be the increase in costs associated with administering the wage incentive system—increased work-in-process inventories, more complicated and detailed accounting records, more grievances over standards, and so on. Hence, under the artificial circumstances that have been assumed, the net benefits credited to the use of wage incentives may be slight. In actual practice the conditions assumed do not hold. Increased efficiency in the use of labor will tend to result in an increase in the volume of the plant's production. Hence, it tends to spread fixed and semifixed costs and to decrease total unit costs.23 If fixed costs per unit are large, the benefit of increased output may be significant. T h u s , the assumption that increased output will lead to increased profits (or decreased losses) is primarily valid. It represents the prevailing tendency that exists in most modern factories. T h e degree to which it is realized depends on: (i) the amount of increase in production; (2) the relation of fixed and semifixed costs to total unit cost; (3) 2 3 All of this may be simply stated in economic terms: T h e increase in the rate of output will lead to a reduction in marginal costs. Hence, if the demand for the firm's product is elastic, the firm will reach a new equilibrium position with a gTeater volume of production; and if the elasticity of demand for the firm's product is greater than unity, the firm's profits will increase (as used here " p r o f i t s " refers to the accountant's concept of revenues exceeding costs).

Theory and Logic of Wage Incentives

93

the costs incurred in installing and administering the wage incentive system; and (4) the elasticity of the demand f o r the firm's products. In summary: T h e worker motivation theory of wage incentives, when applied to factory work, holds that under specified assumptions wage incentives motivate workers to increase their rate of o u t p u t and, in so doing, bring about lower unit costs and a resulting increase in the firm's profits (or decrease of its losses). W h e n analyzed in terms of its underlying assumptions, the theory proves to be consistent and valid; that is, if work is measurable, if the units of measurement are such that there is a direct causal relationship between worker effort and diligence and measured output, if standards of production can be efficiently determined and maintained, if workers are motivated to maximize their monetary income, and if the firm can reduce unit costs by increasing worker output, then wage incentives will contribute to m a x i m i z i n g profits (or minimizing losses). However, analysis of these assumptions indicates that they are not descriptive of all of the real w r orld. N o t all work can be measured in units w h i c h relate worker effort to output. O f t e n standards of performance are set by the use of power and persuasion. W o r k e r s ' motives can conflict. Workers' responses to financial reward may encourage them to falsify production records, mislead time study men, and bargain for loose standards. In short, the worker motivation theory of wage incentives is limited in its application to the payment of rank-and-file workers in manufacturing plants. WAGE

I N C E N T I V E S AS A D E V I C E

MANAGEMENT

T O BE M O R E

FOR

STIMULATING

EFFICIENT

It has previously been indicated that there are two theories j u s t i f y i n g the use of wage incentives. T h e worker mo-

94

Theory and Logic of Wage Incentives

tivation theory has been considered in the preceding section. It remains to analyze the management efficiency theory. Actually these are not separate and distinct. T h e latter includes the former. T h e separation of the two is simply an aid to analysis. It distinguishes the main concepts emphasized by writers and practitioners. T h e management efficiency theory holds that wage incentives not only directly motivate workers but also stimulate management to be more efficient in performing many of its functions, such as determining the best methods of work; standardizing job conditions; setting performance standards; selecting, placing, and training workers; planning and scheduling production; and building teamwork. Implicit in this theory are two basic assumptions: (1) that under normal operating conditions management is relatively inefficient, and (2) that wage incentives will actually stimulate management to do a better job. It is obvious that if management is functioning efficiently it has little to gain from the use of wage incentives. However, in actual practice plants seldom reach perfection. Some improvement is always possible. The point to be emphasized is that the potential gain in this area is inversely related to the existing level of management efficiency. T h e poorer the management, the greater the potential improvement. 24 T h e general nature of the management incentive theory of wage incentives has been indicated." It remains to analyze the impact of wage incentives with respect to setting standards of performance; determining the best method of doing work; planning and scheduling production; and A corollary to this statement it that these plants (the inefficiently managed) are precisely those that are most likely to be unsuccessful in administering wage incentives. Hence, wage incentives are most needed where they are least likely to succeed. See pp. 75-77 above.

T h e o r y and L o g i c of W a g e Incentives

95

recruiting, selecting, and training workers and building teamwork. THE

IMPACT

OF

WAGE

INCENTIVES

ON

THE

DETERMINA-

TION OF PERFORMANCE STANDARDS. O n e of the essential jobs of management is the establishment of standards of performance. Such standards serve as a guide for planning and a basis for evaluating performance. T h e management motivation theory holds that the use of wage incentives forces management to do a better job of setting standards, because standards of performance are essential to a wage incentive plan, and they are explicitly related to worker earnings. Hence, it is argued that under wage incentives standards must be set, and they must be set correctly. There can be no doubt that the use of wage incentives focuses attention on the establishment of performance standards. However, there are reasons to believe that it also tends to inject bias into these standards. This follows from the fact that there is no final and absolute way in which to judge the accuracy of standards. Standards work because workers make them work. 28 If workers do not accept a standard, they may apply a variety of techniques to have it changed (such as slowdowns, strikes, and sabotage). Thus, if management sets tight standards of output, the workers may agitate to have the 2« T h i s fact has been pointed out by numerous students. For example. Mundel says, " M a n y standard times function well, even though they are incorrect, because the workers have learned that it is advantageous to have them function." Systematic Time and Motion Study, p. 156. R a l p h Presgrave, W i l l i a m W h y t e , Clinton Golden, and Harold Ruttenberg have pointed out that (1) many standards may be in error by plus or minus 50 percent; (2) numerous worker grievances are for relaxation of standards; and (3) rechecking of standards reveals that many of them are in error. C. Golden and H. Ruttenberg, The Dynamics of Industrial Democracy (New York, H a r p e r and Brothers, 1942); Presgrave, Dynamics of Time Study; W i l l i a m F. W h y t e , "Incentives for Productivity: T h e B u n d y T u b i n g Case," Applied Anthropology (now Human Organization), Vol. V I I (Spring, 1948), and included in Human Factors in Management (rev. ed.; New York, Harper and Brothers, 1951), ed. by S. D. Hoslett.

96

T h e o r y and Logic of Wage Incentives

rates liberalized. However, if management sets loose rates, it is unlikely that pressures will be exerted to correct the situation. In such instances, it is probable that the workers will set ceilings on production to prevent management from learning of the looseness of its rates. In this manner, under wage incentives, an upward bias (that is, a tendency toward loose standards) is injected into rate setting. T h e special interests of the lower-level supervision may reinforce this bias. Foremen tend to be judged by the bonus earnings of their subordinates. T h e foreman knows that any sudden or significant change in the level of bonus earned in his department will bring about an investigation by his superiors. Hence the foreman's interest is in relatively consistent bonus earnings for his men. One of the easiest ways for him to achieve this end is to secure reasonably loose standards of output and then to enforce ceilings on earnings. This means that when work is flowing smoothly, the rank-and-file workers will maintain a relaxed pace and will significantly restrict output. However, if unusual delays arise, the workers will still be able to maintain their average earnings without undue effort. T H E I M P A C T OF W A G E INCENTIVES ON DETERMINATION OF

Before a performance standard may be set, a definite method for doing the work must be established. T h e question to be answered is: "Do wage incentives stimulate management to do a better job in determining work methods?" Although a categorical answer to this question can not be given, several significant tendencies are noticeable. On the positive side is the fact that if management is to do a good job of setting standards, it must first thoroughly study methods. Hence, in the sense that wage incentives focus attention on methods of work, it can be argued that wage incentives stimulate management to improve work methods. WORK METHODS.

Theory and Logic of Wage Incentives

97

However, the effectiveness of wage incentives must be judged in terms of a definite point of reference. Probably the most realistic comparison to make is between a motion study program under wage incentives and one under a time wage system. When analyzed in this context, it becomes apparent that wage incentives create self-limiting forces. First, wage incentives develop pressures for rapid standard setting. One of the dilemmas of administering wage incentives is determination of how long a new job should run before a standard of performance is established. Until a standard is set, the workers will slow down. They will restrict production in order to encourage the setting of a liberal standard. T h e excessive labor costs arising from such a situation tempt management to apply pressure on the time study men to rate new jobs rapidly. Furthermore, if the workers suffer a loss in earnings while working on an unrated job, added pressure will be applied. Thus, on a new job, pressure tends to be exerted to install rates before the job is running smoothly. T h e time study men need not be too concerned about this fact. They may protect themselves from errors by guaranteeing standards only so long as the method of doing the job is unchanged. Hence, if a loose rate develops, the time study men usually have the excuse that the method was changed.27 A further complication is that wage incentives increase the probability that method changes will lead to friction and unrest. When a method is changed, "proper" standards of performance must be set for the new job. Conflicting values as to what is proper lead to costly grievances. One of the patterns observed in interviews with industrial engineers was the tendency to blame loose rates on the foremen. T h e common statement was: "Loose rates arose from poor foremanship. Foremen failed to report method changes; hence, the precedent of applying the old rate was established."

98

T h e o r y and Logic of Wage Incentives

T h e workers generally use past output as a basis of comparison in determining the "proper" rate. They reason that their pay per unit of output should not be reduced. On the other hand, management argues that worker effort and skill should determine the rate. For example: If management installs a new machine that doubles the worker's rate of output without requiring increased skill or effort on his part, management will maintain that the worker's earnings on the new job should be no greater than they were with the old method. T h e worker may argue that he is doubling his output and should therefore receive double the pay. T h e worker tends to label any standard which falls short of this a "rate cut." In actual practice the worker may accept less than double earnings, but the fact remains that a difference of opinion over standards usually exists. T h e wage incentive system intensifies this difference by directly relating it to the worker's money income. T h e friction encountered in making methods improvements on wage incentive jobs tends to discourage management from making all but the most dramatic and profitable changes. In other words, wage incentives increase the cost of change and make it uneconomical for management to make innovations that might be profitable under a time-wage system. A further limitation of wage incentives as a device for stimulating the development of the best work method arises from the usual guarantee that "rates will remain unchanged so long as methods are constant." T o the worker this implies that any method improvement he devises may lead to a cut in his rate. Thus, the self-interest of the worker discourages him from suggesting major improvements. 28 28 T o overcome this weakness some companies establish special systems to encourage rank-and-file workers to make suggestions. Usually such "sug-

T h e o r y and Logic of W a g e Incentives

g9

Frequently the workers do improve job methods, but the fear of a rate cut may cause them to use numerous ruses to prevent the industrial engineers from learning of the change (for example, they restrict output, hide special tools in their lockers, and the like). Thus, wage incentives tend to prevent the plant from obtaining the full benefit of the worker's knowledge. Wage incentives may also lead management to make "uneconomical" methods changes. This occurs where loose rates have resulted in exorbitant earnings. In order to restore order to the hierarchy of earnings and eliminate excessive labor costs, management may initiate methods changes aimed at justifying a rate cut rather than increasing output. Such methods changes are uneconomical in the sense that they would not have been worth while prior to the establishment of wage incentives. Thus, it appears that management will gain more from a consciously planned program of work simplification and methods improvement under a time rate system of remuneration than it will when such a program is tied to the administration of wage incentives. T h e very use of wage incentives tends to set into motion forces which limit the development of the most efficient methods of work. WAGE INCENTIVES AS A DEVICE FOR AIDING

MANAGEMENT

In the usual manufacturing plant, with its division of labor and specialization of operations, the coordination of production in-

TO PLAN A N D SCHEDULE PRODUCTION.

gestion plans" are administered by a special committee which reviews the suggestions for methods improvement submitted by the workers and then decides on a monetary reward for each cost-saving suggestion that is adopted. T h e field studies of this investigation indicated that unless such suggestion plans are accompanied by a high-pressure propaganda campaign they tend to cause more trouble than they are worth. T h e difficulties center on the questions: Has the person making the suggestion really originated it? Is the reward for a good suggestion adequate or fair? and Why is the processing of suggestions so time consuming?

íoo

Theory and Logic of Wage Incentives

volves detailed scheduling, dispatching, and follow-up. T o minimize costs, management must maintain m i n i m u m inventories and at the same time have adequate supplies on hand. It is argued that wage incentives contribute to management's success in performing these functions by: (1) supplying management with the detailed information required for proper scheduling of production, (2) bringing into focus any unusual variations in job conditions, and (3) encouraging " o p t i m u m " output. Wage incentives as a source of scheduling information. If wage incentives are to function, explicit standards of o u t p u t must be stated and comparisons must be made between standard output and actual output. Hence, as a result of the use of wage incentives, management develops an administrative point of reference for scheduling production and for evaluating performance. 2 9 Wage incentives as a device for bringing into focus serious variations from standard conditions. Standards of performance are predicated upon reasonably standard worki n g conditions. If there are sudden and wide variations f r o m standard conditions, rates become exceptionally tight or loose. From this it is reasoned that wage incentives will automatically bring to management's attention significant deviations from standard conditions. H e r e again a bias arises. Workers will complain only about those changes which limit their rate of output (and hence decrease their earnings). For example, they will file grievances over loss of earnings arising f r o m such events as machine breakdowns, lack of raw materials, and poor quality of raw materials, but they will seldom call to management's attention changes that facilitate production. In fact, if such changes do occur the workers tend to ap2» It should be pointed out that this procedure can also exist independent of wage incentives.

Theory and Logic of Wage Incentives

101

propriate the benefits by relaxing their work pace and/or slightly increasing their output. Seldom do they respond by significantly increasing their rates of output. 30 Wage incentives as a device for encouraging "optimum" output. O n e of the problems in understanding wage incentives arises from a failure to differentiate between the " m a x i m u m " output of workers and their " o p t i m u m " output. M a n y writers imply that workers should always release their m a x i m u m effort and skill to produce as rapidly as possible on every job. In actuality, what is necessary for efficient production is that workers be encouraged to work at an " o p t i m u m " rate. Here " o p t i m u m " refers to the highest rate of worker output that is consistent with coordination of the outputs of the plant's component parts. T h e difference between " o p t i m u m " and " m a x i m u m " o u t p u t becomes apparent in situations where significant variations in work conditions occur. If the worker maximizes his output, his performance will closely correlate with changes in j o b conditions. T h e worker's rate of output will vary widely. T h e frequent fluctuations will tend to disrupt schedules and to necessitate excessive work-inprocess inventories. In contrast, if the worker maintains an " o p t i m u m " rate of output, he will produce at a rate which facilitates the smooth flow of materials through the manufacturing processes. T h e o p t i m u m output of the individual worker is at a level consistent with the smooth coordination of mutually related jobs. T h e point to be emphasized is that the o p t i m u m output includes a concept of steady and predictable performance. W a g e incentives function to promote o p t i m u m output. T h e y provide workers with explicit points of reference in setting production targets and provide workers with spe30 F o r the logic s u p p o r t i n g t h e m a i n t e n a n c e of ceilings on o u t p u t , see pp. 41-43, above.

io2

T h e o r y a n d L o g i c of W a g e

Incentives

cific goals. T h e manner in which this occurs had been described in the section dealing with ceilings on output. 3 1 Many writers fail to see the importance of ceilings on output. T h e y imply that this phenomenon is ipso facto undesirable. T h e weakness of such logic is obvious. Without some sort of ceiling on output, the usual manufacturing plant would find planning and scheduling an extremely difficult and costly task. Day-to-day variations in the same worker, between workers, in raw materials, and in machines would tend to disrupt production schedules. Thus, in the usual manufacturing plant, with its specialization and division of labor, it is apparent that ceilings on output encouraged by the use of wage incentives may serve a useful function. T h e y may allow closer scheduling and better coordination of the production processes. T h e net benefit that may be attributed to wage incentives in this stabilizing of rates of output is hard to evaluate. It depends on (1) the level of output at which ceilings are set, (2) the degree to which workers can maintain earnings by subterfuge or special allowances, and (3) the extent to which workers build up "kitties" of unreported production. If standards are so loose that ceiling on output are set at a level that leaves workers with excessive idle time, the loss due to excessive labor costs may offset the gain arising from a more predictable rate of output. In addition, if workers maintain their earnings at a relatively stable level by cheating or winning special allowances, the actual level of output will not be constant. Furthermore, if workers build up excessive inventories of semiprocessed parts ("kitties"), the relatively stable rates of production as reported through the wage incentive system will be a fiction. In summary, a central problem in plant management is the coordination of the various mutually related parts 31

See pp. 41-43, above.

Theory and Logic of Wage Incentives

103

of the operating unit. T h i s is facilitated by "optimum" production. Wage incentives, by encouraging ceilings on output, contribute to the stability and predictability of production. T h e degree to which this results in optimum production depends on the level of worker efficiency at which ceilings are set and the extent to which workers' earnings correlate with their actual rates of output. THE I M P A C T OF WAGE INCENTIVES ON PERSONNEL MANAGE-

One of the justifications for the use of wage incentives is that they aid management in dealing with rankand-file personnel. T h i s has been suggested by numerous writers and practitioners. T h e points of emphasis vary. T h e general tenor of the remarks may be discussed in terms of the impact of wage incentives on recruitment, selection, placement, and training of workers, and on cooperative relations in the plant. Wage incentives as a device for attracting and holding workers. It is pointed out that earnings are higher under wage incentives than they are under time wage systems.32 Hence, it is argued that wage incentives are a device for attracting and holding workers. T h e assumptions implied in this reasoning are that (1) workers know of the earnings on alternative jobs, and (2) workers are attracted to those jobs providing the greatest earnings. T h e importance of wages in attracting labor has been the subject of numerous empirical studies. 33 T h e general consensus of their findings is: MENT.

See pp. 22-24, above. 38 As examples of such studies, see W. Rupert Malaurin and Charles A. Myers, "Wages and the Movement of Factory Labor," The Quarterly Journal of Economics, LVII (February, 1943), 241-64; Lloyd George Reynolds and Joseph Shister, Job Horizons (New York, Harper and Brothers, 1948), Chaps. I l l and IV, pp. 84, 87-88; E. Wight Bakke, The Unemployed Worker (New Haven, Yale University Press, 1940), pp. 165-252; Reynolds, Structure of Labor Markets, Chaps. VIII and IX; Charles A. Myers and George P. Shultz, The Dynamics of a Labor Market (New York, PrenticeHall, Inc., 1951), Chap. IV. 32

104

Theory and Logic of Wage Incentives

1. Rank-and-file workers seldom have detailed knowledge of job opportunities. Friends and relatives are the principal sources of information. 2. Rank-and-file workers seldom systematically canvass the market. T h e y have only fragmentary information about job openings, rates of pay, and the like. 3. Most moves to new jobs are forced by layoffs or discharges. 4. Only in the case of voluntary job movement is there a trend toward the higher-wage firms. T h e inference to be drawn from existing field research is that wages are not of major importance in attracting labor to a given firm. Only when frequent voluntary job movements exist will high wages be a significant factor in recruiting rank-and-file employees. T h e role of wages in holding employees on given employment has been a subject of controversy. Numerous attitude surveys have been made to determine the relative rank of wages in contributing to the worker's j o b satisfaction. T h e findings of such studies generally reveal that wages rank relatively low (from third to seventh). 34 It is questionable whether these surveys contribute to an understanding of the role of wages in reducing turnover. J o b satisfaction is not composed of many separate elements. It is an alloy. T o ask which is more important s * For example: A study made in 1935 by Hoppock and Spiegler revealed that in a small work group the factors most frequently mentioned as important to job satisfaction were, in order of importance: (1) work associates, (2) the work itself, (3) the boss, (4) variety, (5) freedom in work, (6) hours, and (7) earnings. R . Hoppock and S. Spiegler, " J o b Satisfaction," Occupations, The Vocational Guidance Magazine, X V I (1938), 636-43. Also see: R . B. Hersey, "Psychology of Workers," Personnel Journal, X I V (1936), 291-96; National Industrial Conference Board, Factors Affecting Employee Morale, in Studies in Personnel Policy, No. 85 (1947); Lawrence G. Lindahl, "What Makes a Good Job?" Personnel, X X V (January, 1949), 263-66; M. L. Blum and J . J . Russ, "A Study of Employee Attitudes toward Various Incentives," Personnel, X I X (1942), 438-44; Clifford Jurgensen, "Selected Factors Which Influence J o b Performance," Journal of Applied Psychology, X X X I (1937), 553-64.

Theory and Logic of Wage Incentives

105

to the worker—human relations on the job (independence, fairness of treatment, and the like), the intrinsic nature of the job (skills, variety, conditions of work, and the like), or wages—is like asking which leg of a three-legged stool is most important. T h e factors entering into job satisfaction are not separate and independent. T h e y exist in a system of mutual causality, a system in which everything depends on everything else. Despite the empirical evidence that, generally, wages are unimportant in attracting labor, experience in the Second W o r l d W a r indicated that in a tight labor market relative wages are important. During this period many firms adopted wage incentives as a means of circumventing wage stabilization regulations. 3 5 By setting loose standards and granting liberal allowances, these plants significantly increased the earnings of their workers. 36 In summary, there is little evidence that the high earnings associated with wage incentives will always be effective in attracting and holding rank-and-file factory work35 In July, 1942, the National War Labor Board announced its policy of stabilizing wages. It publicized its Little Steel Formula providing that wages were stabilized at a level 15 percent above that of January 1, 1941. On October 3, 1942, Executive Order No. 9250 was issued. It provided that: " T h e National W a r Labor Board shall not approve any increase in the wage rates prevailing on September 15, 1942, unless such increase is necessary to correct maladjustments or inequalities, to eliminate substandards of living, to correct gross inequities, or to aid in the effective prosecution of the war." Steel Case Research Committee of the Steel Industry, Steel Wages, a National Issue (April, 1944), pp. 5, 8. 3« T h e only wage control over the operation of an existing incentive plan was the general requirement that, with changes in products or methods, new standards must be set "in line" with existing rates. T h i s control breaks down under conditions comparable to those of World War II. Such things as increased product runs and more standardized products contributed to increased output. Such conditions also created the opportunity for a general loosening of production standards. " T h a t this loosening of production standards took place quite extensively in the picce rate industries and in other incentive operations during World War II is hardly open to serious debate." E. Robert Livernash, "Stabilization of the Internal Wage R a t e Structure," Industrial and Labor Relations Review, VII (January. >954). 216.

io6

Theory and Logic of Wage Incentives

ers. However, under special conditions, such as those existing in the labor market of the Second World War, there is reason to believe that wage incentives may be important in recruiting labor and reducing turnover. 37 The impact of wage incentives on selection and placement of workers. One of the important management problems arising from the existence of individual differences and specialization is the placing of individuals on jobs that are compatible with their abilities, interests, and ambitions. With respect to this, the job of management is to differentiate among workers so as to place each individual in the j o b that will allow him to make his greatest contribution to enterprise profits. Worker contributions to enterprise profit are of two types. 38 1. Minimizing the cost of producing a given output. T h i s involves economizing in the use of the materials, equipment, and the worker's time. 2. Coordinating the component parts of the manufacturing plant. T h a t is, the worker's contribution to cooperative relationships within the plant. Wage incentives are one of the devices used by management for differentiating between workers. T h e y are a means of directly discriminating between workers on the basis of output. T h e y automatically relate earnings to operator efficiency as judged by an explicit standard of In the cases covered by this investigation there was a noticeable tendency for wage incentives to result in evasion of wage stabilization regulations. However, it should be pointed out that not all of the managements were fully conscious of what they were doing. Frequently they reported, and evidence bore them out, that they settled each wage incentive grievance on an ad hoc basis. T h e primary explanation for deterioration of standards lies in the general economic forces operating during the World War II labor market. T h e films were more interested in getting out the production than in cutting costs. T h e excess profits tax and cost plus contracts encouraged this attitude. Furthermore, the acute scarcity of labor provided workers with increased power in bargaining over standards. 38 T . L. Whisler, Merit Rating: A Management T o o l (unpublished Ph.D. dissertation, University of Chicago, School of Business, March, 1953), p. 79.

Theory and Logic of Wage Incentives

107

performance. Thus, they are an aid in identifying the strongly motivated workers. In so doing, they focus management attention on variations in performance and provide management with a guide for promoting, transferring, demoting, and discharging workers. 39 However, there are limitations implied by the preceding discussion: 1. T h e appropriateness of differentiations resulting from the use of wage incentives will be limited by the accuracy of the standard used in determining worker efficiency. T h e tendency for standards to be inconsistent and inaccurate (as judged by the relationship of skill and effort to earnings) implies that differences in workers' earnings may be caused by weaknesses in setting correct standards rather than differences in the diligence and effort of the various individuals. 2. Differentiation among workers on the basis of wage incentive earnings is not a complete measure of their respective contributions to enterprise profits. It fails to consider the worker's contribution to the coordination of the plant's various activities. In fact, the individuals earning the most, the "rate busters," tend to detract from harmonious relations between workers. Frequently, rate busters cause so much friction that it is unrealistic to judge them solely in terms of their output. These weaknesses in the use of wage incentives as a means of differentiating between workers lead to the conclusion that, although they may be helpful in this process, they should always be supplemented by other devices, and should never be considered completely reliable. The impact of wage incentives on the training of rankand-file workers. Wage incentives may be effective in bringing to light training needs. For example: If one or two 39 A few of the other devices used for selecting and placing workers are interviews, merit ratings, and work tests.

io8

T h e o r y and Logic of Wage Incentives

workers in a work group consistently fall below the expected rate of output, their low earnings record under the wage incentive system may prompt management to investigate. T h e investigation may reveal a training problem rather than tight standards or lack of motivation. In addition, the mechanics of administering wage incentives usually provide workers with detailed information on the proper method of performing the work. T h e description of the work method may serve as a job instruction device. The impact of wage incentives on cooperative relations in the plant. At the rank-and-file level, wage incentives produce both desirable and undesirable effects on cooperation. On the positive side, they may integrate the interests of workers by focusing their attention on the increased earnings that result from cooperating to attain increased output. On the negative side are numerous problems growing out of the difficulties surrounding the setting and maintaining of accurate and consistent standards of performance. Pigors and Myers have indicated the general nature of this problem. They point out: "Wage differentials are a mark of social status in the factory organization. If they do not correspond with the relative significance of jobs as employees view them, the workers' sense of justice is outraged." 40 Illogical and dynamic variations in earnings injected into the plant wage structure by wage incentives tend to engender serious enmities among workers. Jobs that normally rate low in status may pay better than the traditionally "good" jobs. Hence, workers holding the "good" jobs may feel that established differentials should be restored. They will apply pressure to have management loosen «o p. pigors and C. Myers, Personnel Administration McGraw-Hill Book Co., Inc., 1950), p. 255.

(2d c